Brain tumors,one of the most lethal diseases with low survival rates,require early detection and accurate diagnosis to enable effective treatment planning.While deep learning architectures,particularly Convolutional N...Brain tumors,one of the most lethal diseases with low survival rates,require early detection and accurate diagnosis to enable effective treatment planning.While deep learning architectures,particularly Convolutional Neural Networks(CNNs),have shown significant performance improvements over traditional methods,they struggle to capture the subtle pathological variations between different brain tumor types.Recent attention-based models have attempted to address this by focusing on global features,but they come with high computational costs.To address these challenges,this paper introduces a novel parallel architecture,ParMamba,which uniquely integrates Convolutional Attention Patch Embedding(CAPE)and the Conv Mamba block including CNN,Mamba and the channel enhancement module,marking a significant advancement in the field.The unique design of ConvMamba block enhances the ability of model to capture both local features and long-range dependencies,improving the detection of subtle differences between tumor types.The channel enhancement module refines feature interactions across channels.Additionally,CAPE is employed as a downsampling layer that extracts both local and global features,further improving classification accuracy.Experimental results on two publicly available brain tumor datasets demonstrate that ParMamba achieves classification accuracies of 99.62%and 99.35%,outperforming existing methods.Notably,ParMamba surpasses vision transformers(ViT)by 1.37%in accuracy,with a throughput improvement of over 30%.These results demonstrate that ParMamba delivers superior performance while operating faster than traditional attention-based methods.展开更多
Detection of brain tumors in MRI images is the first step in brain cancer diagnosis.The accuracy of the diagnosis depends highly on the expertise of radiologists.Therefore,automated diagnosis of brain cancer from MRI ...Detection of brain tumors in MRI images is the first step in brain cancer diagnosis.The accuracy of the diagnosis depends highly on the expertise of radiologists.Therefore,automated diagnosis of brain cancer from MRI is receiving a large amount of attention.Also,MRI tumor detection is usually followed by a biopsy(an invasive procedure),which is a medical procedure for brain tumor classification.It is of high importance to devise automated methods to aid radiologists in brain cancer tumor diagnosis without resorting to invasive procedures.Convolutional neural network(CNN)is deemed to be one of the best machine learning algorithms to achieve high-accuracy results in tumor identification and classification.In this paper,a CNN-based technique for brain tumor classification has been developed.The proposed CNN can distinguish between normal(no-cancer),astrocytoma tumors,gliomatosis cerebri tumors,and glioblastoma tumors.The implemented CNN was tested on MRI images that underwent a motion-correction procedure.The CNN was evaluated using two performance measurement procedures.The first one is a k-fold cross-validation testing method,in which we tested the dataset using k=8,10,12,and 14.The best accuracy for this procedure was 96.26%when k=10.To overcome the over-fitting problem that could be occurred in the k-fold testing method,we used a hold-out testing method as a second evaluation procedure.The results of this procedure succeeded in attaining 97.8%accuracy,with a specificity of 99.2%and a sensitivity of 97.32%.With this high accuracy,the developed CNN architecture could be considered an effective automated diagnosis method for the classification of brain tumors from MRI images.展开更多
Knowledge-based transfer learning techniques have shown good performance for brain tumor classification,especially with small datasets.However,to obtain an optimized model for targeted brain tumor classification,it is...Knowledge-based transfer learning techniques have shown good performance for brain tumor classification,especially with small datasets.However,to obtain an optimized model for targeted brain tumor classification,it is challenging to select a pre-trained deep learning(DL)model,optimal values of hyperparameters,and optimization algorithm(solver).This paper first presents a brief review of recent literature related to brain tumor classification.Secondly,a robust framework for implementing the transfer learning technique is proposed.In the proposed framework,a Cartesian product matrix is generated to determine the optimal values of the two important hyperparameters:batch size and learning rate.An extensive exercise consisting of 435 simulations for 11 state-of-the-art pre-trained DL models was performed using 16 paired hyperparameters from the Cartesian product matrix to input the model with the three most popular solvers(stochastic gradient descent with momentum(SGDM),adaptive moment estimation(ADAM),and root mean squared propagation(RMSProp)).The 16 pairs were formed using individual hyperparameter values taken from literature,which generally addressed only one hyperparameter for optimization,rather than making a grid for a particular range.The proposed framework was assessed using a multi-class publicly available dataset consisting of glioma,meningioma,and pituitary tumors.Performance assessment shows that ResNet18 outperforms all other models in terms of accuracy,precision,specificity,and recall(sensitivity).The results are also compared with existing state-of-the-art research work that used the same dataset.The comparison was mainly based on performance metric“accuracy”with support of three other parameters“precision,”“recall,”and“specificity.”The comparison shows that the transfer learning technique,implemented through our proposed framework for brain tumor classification,outperformed all existing approaches.To the best of our knowledge,the proposed framework is an efficient framework that helped reduce the computational complexity and the time to attain optimal values of two important hyperparameters and consequently the optimized model with an accuracy of 99.56%.展开更多
The deep learning models are identified as having a significant impact on various problems.The same can be adapted to the problem of brain tumor classification.However,several deep learning models are presented earlie...The deep learning models are identified as having a significant impact on various problems.The same can be adapted to the problem of brain tumor classification.However,several deep learning models are presented earlier,but they need better classification accuracy.An efficient Multi-Feature Approximation Based Convolution Neural Network(CNN)model(MFACNN)is proposed to handle this issue.The method reads the input 3D Magnetic Resonance Imaging(MRI)images and applies Gabor filters at multiple levels.The noise-removed image has been equalized for its quality by using histogram equalization.Further,the features like white mass,grey mass,texture,and shape are extracted from the images.Extracted features are trained with deep learning Convolution Neural Network(CNN).The network has been designed with a single convolution layer towards dimensionality reduction.The texture features obtained from the brain image have been transformed into a multi-dimensional feature matrix,which has been transformed into a single-dimensional feature vector at the convolution layer.The neurons of the intermediate layer are designed to measure White Mass Texture Support(WMTS),GrayMass Texture Support(GMTS),WhiteMass Covariance Support(WMCS),GrayMass Covariance Support(GMCS),and Class Texture Adhesive Support(CTAS).In the test phase,the neurons at the intermediate layer compute the support as mentioned above values towards various classes of images.Based on that,the method adds a Multi-Variate Feature Similarity Measure(MVFSM).Based on the importance ofMVFSM,the process finds the class of brain image given and produces an efficient result.展开更多
Deep learning(DL)techniques,which do not need complex preprocessing and feature analysis,are used in many areas of medicine and achieve promising results.On the other hand,in medical studies,a limited dataset decrease...Deep learning(DL)techniques,which do not need complex preprocessing and feature analysis,are used in many areas of medicine and achieve promising results.On the other hand,in medical studies,a limited dataset decreases the abstraction ability of the DL model.In this context,we aimed to produce synthetic brain images including three tumor types(glioma,meningioma,and pituitary),unlike traditional data augmentation methods,and classify them with DL.This study proposes a tumor classification model consisting of a Dense Convolutional Network(DenseNet121)-based DL model to prevent forgetting problems in deep networks and delay information flow between layers.By comparing models trained on two different datasets,we demonstrated the effect of synthetic images generated by Cycle Generative Adversarial Network(CycleGAN)on the generalization of DL.One model is trained only on the original dataset,while the other is trained on the combined dataset of synthetic and original images.Synthetic data generated by CycleGAN improved the best accuracy values for glioma,meningioma,and pituitary tumor classes from 0.9633,0.9569,and 0.9904 to 0.9968,0.9920,and 0.9952,respectively.The developed model using synthetic data obtained a higher accuracy value than the related studies in the literature.Additionally,except for pixel-level and affine transform data augmentation,synthetic data has been generated in the figshare brain dataset for the first time.展开更多
Background Brain tumors are challenging to diagnose and treat,and require accurate and early therapeutic intervention.Magnetic Resonance Imaging(MRI)scans can visualize the internal structure of the brain.Often,deep l...Background Brain tumors are challenging to diagnose and treat,and require accurate and early therapeutic intervention.Magnetic Resonance Imaging(MRI)scans can visualize the internal structure of the brain.Often,deep learning is applied to images for the early and accurate detection of tumor cells.However,these models lack accuracy and efficacy in practical applications.Hybrid or modified models can facilitate better classification and provide insights into early-stage cancer detection.Methods This study demonstrates a parallel architecture that uses MRI images and integrates transformer-based frameworks with Convolutional Neural Networks(CNNs)to better classify distinct types of brain tumors.The proposed architecture,SwinResDual(SwRD),combines a Residual Network(ResNet)and a Swin Transformer in parallel to extract key features from input images.Using augmented MRI scans,31,464 scans for multiclass classification,and 30,000 scans for binary classification,the architecture simultaneously processed images through the ResNet50 and Swin Transformer branches,leveraging their strengths in hierarchical feature extraction and global context modeling to efficiently capture local and global image features.The final classifications are obtained by merging these features and passing them through a linear classifier.This approach identifies strong and varied characteristics and provides a precise brain tumor diagnosis.Results In the extensive evaluation,the model performed with an accuracy of 99.79% and a cross-validation accuracy of 100%for multiclass classification,along with 99.97% accuracy in binary classification.Conclusions In conclusion,the findings demonstrate great promise for brain tumor detection and advanced medical imaging diagnostics.展开更多
Brain tumor classification is crucial for personalized treatment planning.Although deep learning-based Artificial Intelligence(AI)models can automatically analyze tumor images,fine details of small tumor regions may b...Brain tumor classification is crucial for personalized treatment planning.Although deep learning-based Artificial Intelligence(AI)models can automatically analyze tumor images,fine details of small tumor regions may be overlooked during global feature extraction.Therefore,we propose a brain tumor Magnetic Resonance Imaging(MRI)classification model based on a global-local parallel dual-branch structure.The global branch employs ResNet50 with a Multi-Head Self-Attention(MHSA)to capture global contextual information from whole brain images,while the local branch utilizes VGG16 to extract fine-grained features from segmented brain tumor regions.The features from both branches are processed through designed attention-enhanced feature fusion module to filter and integrate important features.Additionally,to address sample imbalance in the dataset,we introduce a category attention block to improve the recognition of minority classes.Experimental results indicate that our method achieved a classification accuracy of 98.04%and a micro-average Area Under the Curve(AUC)of 0.989 in the classification of three types of brain tumors,surpassing several existing pre-trained Convolutional Neural Network(CNN)models.Additionally,feature interpretability analysis validated the effectiveness of the proposed model.This suggests that the method holds significant potential for brain tumor image classification.展开更多
Breast cancer detection heavily relies on medical imaging, particularly ultrasound, for early diagnosis and effectivetreatment. This research addresses the challenges associated with computer-aided diagnosis (CAD) of ...Breast cancer detection heavily relies on medical imaging, particularly ultrasound, for early diagnosis and effectivetreatment. This research addresses the challenges associated with computer-aided diagnosis (CAD) of breastcancer fromultrasound images. The primary challenge is accurately distinguishing between malignant and benigntumors, complicated by factors such as speckle noise, variable image quality, and the need for precise segmentationand classification. The main objective of the research paper is to develop an advanced methodology for breastultrasound image classification, focusing on speckle noise reduction, precise segmentation, feature extraction, andmachine learning-based classification. A unique approach is introduced that combines Enhanced Speckle ReducedAnisotropic Diffusion (SRAD) filters for speckle noise reduction, U-NET-based segmentation, Genetic Algorithm(GA)-based feature selection, and Random Forest and Bagging Tree classifiers, resulting in a novel and efficientmodel. To test and validate the hybrid model, rigorous experimentations were performed and results state thatthe proposed hybrid model achieved accuracy rate of 99.9%, outperforming other existing techniques, and alsosignificantly reducing computational time. This enhanced accuracy, along with improved sensitivity and specificity,makes the proposed hybrid model a valuable addition to CAD systems in breast cancer diagnosis, ultimatelyenhancing diagnostic accuracy in clinical applications.展开更多
Liver cancer remains a leading cause of mortality worldwide,and precise diagnostic tools are essential for effective treatment planning.Liver Tumors(LTs)vary significantly in size,shape,and location,and can present wi...Liver cancer remains a leading cause of mortality worldwide,and precise diagnostic tools are essential for effective treatment planning.Liver Tumors(LTs)vary significantly in size,shape,and location,and can present with tissues of similar intensities,making automatically segmenting and classifying LTs from abdominal tomography images crucial and challenging.This review examines recent advancements in Liver Segmentation(LS)and Tumor Segmentation(TS)algorithms,highlighting their strengths and limitations regarding precision,automation,and resilience.Performance metrics are utilized to assess key detection algorithms and analytical methods,emphasizing their effectiveness and relevance in clinical contexts.The review also addresses ongoing challenges in liver tumor segmentation and identification,such as managing high variability in patient data and ensuring robustness across different imaging conditions.It suggests directions for future research,with insights into technological advancements that can enhance surgical planning and diagnostic accuracy by comparing popular methods.This paper contributes to a comprehensive understanding of current liver tumor detection techniques,provides a roadmap for future innovations,and improves diagnostic and therapeutic outcomes for liver cancer by integrating recent progress with remaining challenges.展开更多
Accurate cancer staging is the foundation of precision oncology and guides prognosis prediction and therapeutic decision-making. The conjoint TNM System by the American Joint Committee on Cancer (AJCC) and the Interna...Accurate cancer staging is the foundation of precision oncology and guides prognosis prediction and therapeutic decision-making. The conjoint TNM System by the American Joint Committee on Cancer (AJCC) and the International Union Against Cancer (UICC) has served as the global standard for tumor classification since inception.展开更多
The latest edition of the WHO classification of the central nervous system was published in 2021.This review summarizes the major revisions to the classification of anterior pituitary tumors.The most important revisio...The latest edition of the WHO classification of the central nervous system was published in 2021.This review summarizes the major revisions to the classification of anterior pituitary tumors.The most important revision involves preferring the terminology of pituitary neuroendocrine tumor(PitNET),even though the terminology of pituitary adenoma(PA)still can be used according to this WHO classification compared to the previous one.Moreover,immunohistochemistry(IHC)examination of pituitary-specific transcription factors(TFs),including PIT1,TPIT,SF-1,GATA2/3,and ERα,is endorsed to determine the tumor cell lineage and to facilitate the classification of PitNET/PA subgroups.However,TF-negative IHC staining indicates PitNET/PA with no distinct cell lineages,which includes unclassified plurihormonal(PH)tumors and null cell(NC)tumors in this edition.The new WHO classification of PitNET/PA has incorporated tremendous advances in the understanding of the cytogenesis and pathogenesis of pituitary tumors.However,due to the shortcomings of the technology used in the diagnosis of PitNET/PA and the limited understanding of the tumorigenesis of PitNET/PA,the application of this new classification system in practice should be further evaluated and validated.Besides providing information for deciding the follow-up plans and adjunctive treatment after surgery,this classification system offers no additional help for neurosurgeons in clinical practice,especially in determining the treatment strategies.Therefore,it is necessary for neurosurgeons to establish a comprehensive pituitary classification system for PitNET/PA that incorporates neuroimaging grading data or direct observation of invasiveness during operation or the predictor of prognosis,as well as pathological diagnosis,thereby distinguishing the invasiveness of the tumor and facilitating neurosurgeons to decide on the treatment strategies and follow-up plans as well as adjunctive treatment after surgery.展开更多
Image-based breast tumor classification is an active and challenging problem.In this paper,a robust breast tumor classification framework is presented based on deep feature representation learning and exploiting avail...Image-based breast tumor classification is an active and challenging problem.In this paper,a robust breast tumor classification framework is presented based on deep feature representation learning and exploiting available information in existing samples.Feature representation learning of mammograms is fulfilled by a modified nonnegative matrix factorization model called LPML-LRNMF,which is motivated by hierarchical learning and layer-wise pre-training(LP)strategy in deep learning.Low-rank(LR)constraint is integrated into the feature representation learning model by considering the intrinsic characteristics of mammograms.Moreover,the proposed LPML-LRNMF model is optimized via alternating direction method of multipliers and the corresponding convergence is analyzed.For completing classification,an inverse projection sparse representation model is introduced to exploit information embedded in existing samples,especially in test ones.Experiments on the public dataset and actual clinical dataset show that the classification accuracy,specificity and sensitivity achieve the clinical acceptance level.展开更多
Objective To explore classification and surgical approach of magnum foramen tumor. Methods A retrospective analysis was performed for 43 surgically treated patients with tumors involving foramen magnum. According to t...Objective To explore classification and surgical approach of magnum foramen tumor. Methods A retrospective analysis was performed for 43 surgically treated patients with tumors involving foramen magnum. According to the site of tumor,the classification was divided into:Type Ⅰ,located at dorsal,Ⅰ a extra-medullary,展开更多
This study investigated the accuracy of MRI features in differentiating the pathological grades of pancreatic neuroendocrine neoplasms（PNENs）. A total of 31 PNENs patients were retrospectively evaluated, including 1...This study investigated the accuracy of MRI features in differentiating the pathological grades of pancreatic neuroendocrine neoplasms（PNENs）. A total of 31 PNENs patients were retrospectively evaluated, including 19 cases in grade 1, 5 in grade 2, and 7 in grade 3. Plain and contrastenhanced MRI was performed on all patients. MRI features including tumor size, margin, signal intensity, enhancement patterns, degenerative changes, duct dilatation and metastasis were analyzed. Chi square tests, Fisher＇s exact tests, one-way ANOVA and ROC analysis were conducted to assess the associations between MRI features and different tumor grades. It was found that patients with older age, tumors with higher TNM stage and without hormonal syndrome had higher grade of PNETs（all P〈0.05）. Tumor size, shape, margin and growth pattern, tumor pattern, pancreatic and bile duct dilatation and presence of lymphatic and distant metastasis as well as MR enhancement pattern and tumor-topancreas contrast during arterial phase were the key features differentiating tumors of all grades（all P〈0.05）. ROC analysis revealed that the tumor size with threshold of 2.8 cm, irregular shape, pancreatic duct dilatation and lymphadenopathy showed satisfactory sensitivity and specificity in distinguishing grade 3 from grade 1 and grade 2 tumors. Features of peripancreatic tissue or vascular invasion, and distant metastasis showed high specificity but relatively low sensitivity. In conclusion, larger size, poorlydefined margin, heterogeneous enhanced pattern during arterial phase, duct dilatation and the presence of metastases are common features of higher grade PNENs. Plain and contrast-enhanced MRI provides the ability to differentiate tumors with different pathological grades.展开更多
Gene expression microarray data can be used to classify tumor types. We proposed a new procedure to classify human tumor samples based on microarray gene expressions by using a hybrid supervised learning method called...Gene expression microarray data can be used to classify tumor types. We proposed a new procedure to classify human tumor samples based on microarray gene expressions by using a hybrid supervised learning method called MOEA+WV (Multi-Objective Evolutionary Algorithm+Weighted Voting). MOEA is used to search for a relatively few subsets of informative genes from the high-dimensional gene space, and WV is used as a classification tool. This new method has been applied to predicate the subtypes of lymphoma and outcomes of medulloblastoma. The results are relatively accurate and meaningful compared to those from other methods. Key words bioinformatics - tumor classification - Pareto optimization - MOEA CLC number Q 786 - TP 181 Foundation item: Supported by the National Natural Science Foundation of China (60301009), the Foundation of Young Scholars of Ministry of Education of China (150118) and Chenguang Project of Wuhan City (211121009).Biography: Liu Juan (1970-), female, Associate Professor, Postdoctoral, research direction: bioinformatics, data mining, machine learning.展开更多
This systematic review aims to comprehensively examine and compare deep learning methods for brain tumor segmentation and classification using MRI and other imaging modalities,focusing on recent trends from 2022 to 20...This systematic review aims to comprehensively examine and compare deep learning methods for brain tumor segmentation and classification using MRI and other imaging modalities,focusing on recent trends from 2022 to 2025.The primary objective is to evaluate methodological advancements,model performance,dataset usage,and existing challenges in developing clinically robust AI systems.We included peer-reviewed journal articles and highimpact conference papers published between 2022 and 2025,written in English,that proposed or evaluated deep learning methods for brain tumor segmentation and/or classification.Excluded were non-open-access publications,books,and non-English articles.A structured search was conducted across Scopus,Google Scholar,Wiley,and Taylor&Francis,with the last search performed in August 2025.Risk of bias was not formally quantified but considered during full-text screening based on dataset diversity,validation methods,and availability of performance metrics.We used narrative synthesis and tabular benchmarking to compare performance metrics(e.g.,accuracy,Dice score)across model types(CNN,Transformer,Hybrid),imaging modalities,and datasets.A total of 49 studies were included(43 journal articles and 6 conference papers).These studies spanned over 9 public datasets(e.g.,BraTS,Figshare,REMBRANDT,MOLAB)and utilized a range of imaging modalities,predominantly MRI.Hybrid models,especially ResViT and UNetFormer,consistently achieved high performance,with classification accuracy exceeding 98%and segmentation Dice scores above 0.90 across multiple studies.Transformers and hybrid architectures showed increasing adoption post2023.Many studies lacked external validation and were evaluated only on a few benchmark datasets,raising concerns about generalizability and dataset bias.Few studies addressed clinical interpretability or uncertainty quantification.Despite promising results,particularly for hybrid deep learning models,widespread clinical adoption remains limited due to lack of validation,interpretability concerns,and real-world deployment barriers.展开更多
One of the leading causes of mortality worldwide is liver cancer.The earlier the detection of hepatic tumors,the lower the mortality rate.This paper introduces a computer-aided diagnosis system to extract hepatic tumo...One of the leading causes of mortality worldwide is liver cancer.The earlier the detection of hepatic tumors,the lower the mortality rate.This paper introduces a computer-aided diagnosis system to extract hepatic tumors from computed tomography scans and classify them into malignant or benign tumors.Segmenting hepatic tumors from computed tomography scans is considered a challenging task due to the fuzziness in the liver pixel range,intensity values overlap between the liver and neighboring organs,high noise from computed tomography scanner,and large variance in tumors shapes.The proposed method consists of three main stages;liver segmentation using Fast Generalized Fuzzy C-Means,tumor segmentation using dynamic thresholding,and the tumor’s classification into malignant/benign using support vector machines classifier.The performance of the proposed system was evaluated using three liver benchmark datasets,which are MICCAI-Sliver07,LiTS17,and 3Dircadb.The proposed computer adided diagnosis system achieved an average accuracy of 96.75%,sensetivity of 96.38%,specificity of 95.20%and Dice similarity coefficient of 95.13%.展开更多
Lung adenocarcinoma (LUAD) is the most common and deadliest subtype of lung cancer. To select moretargeted and effective treatments for individuals, further advances in classifying LUAD are urgently needed. Thenumber,...Lung adenocarcinoma (LUAD) is the most common and deadliest subtype of lung cancer. To select moretargeted and effective treatments for individuals, further advances in classifying LUAD are urgently needed. Thenumber, type, and function of T cells in the tumor microenvironment (TME) determine the progression andtreatment response of LUAD. Long noncoding RNAs (lncRNAs), may regulate T cell differentiation, development,and activation. Thus, our aim was to identify T cell-related lncRNAs (T cell-Lncs) in LUAD and to investigatewhether T cell-Lncs could serve as potential stratifiers and therapeutic targets. Seven T cell-Lncs were identified tofurther establish the T cell-related lncRNA risk score (TRS) in LUAD. Low TRS individuals were characterized byrobust immune status, fewer genomic alterations, and remarkably longer survival than high TRS individuals. Theexcellent accuracy of TRS in predicting overall survival (OS) was validated in the TCGA-LUAD training cohort andthe GEO-LUAD validation cohort. Our data demonstrated the favorable predictive power of the TRS-basednomogram, which had important clinical significance in estimating the survival probability for individuals. Inaddition, individuals with low TRS could respond better to chemotherapy and immunotherapy than those with highTRS. LINC00525 was identified as a valuable study target, and the ability of LUAD to proliferate or invade wassignificantly attenuated by downregulation of LINC00525. In conclusion, the TRS established by T cell-Lncs couldunambiguously classify LUAD patients, predict their prognosis and guide their management. Moreover, our identifiedT cell-Lncs could provide potential therapeutic targets for LUAD.展开更多
DNA methylation is the most intensively studied epigenetic phenomenon, disturbances of which result in changes ingene transcription, thus exerting drastic imparts onto biological behaviors of cancer. Both the global d...DNA methylation is the most intensively studied epigenetic phenomenon, disturbances of which result in changes ingene transcription, thus exerting drastic imparts onto biological behaviors of cancer. Both the global demethylation andthe local hypermethylation have been widely reported in all types of tumors, providing both challenges and opportunitiesfor a better understanding and eventually controlling of the malignance. However, we are still in the very early stage ofinformation accumulation concerning the tumor associated changes in DNA methylation pattern. A number of excellentrecent reviews have covered this issue in depth. Therefore, this review will summarize our recent data on DNA methy-lation profiling in cancers. Perspectives for the future direction in this dynamic and exciting field will also be given.展开更多
Automatic segmentation and classification of brain tumors are of great importance to clinical treatment.However,they are challenging due to the varied and small morphology of the tumors.In this paper,we propose a mult...Automatic segmentation and classification of brain tumors are of great importance to clinical treatment.However,they are challenging due to the varied and small morphology of the tumors.In this paper,we propose a multitask multiscale residual attention network(MMRAN)to simultaneously solve the problem of accurately segmenting and classifying brain tumors.The proposed MMRAN is based on U-Net,and a parallel branch is added at the end of the encoder as the classification network.First,we propose a novel multiscale residual attention module(MRAM)that can aggregate contextual features and combine channel attention and spatial attention better and add it to the shared parameter layer of MMRAN.Second,we propose a method of dynamic weight training that can improve model performance while minimizing the need for multiple experiments to determine the optimal weights for each task.Finally,prior knowledge of brain tumors is added to the postprocessing of segmented images to further improve the segmentation accuracy.We evaluated MMRAN on a brain tumor data set containing meningioma,glioma,and pituitary tumors.In terms of segmentation performance,our method achieves Dice,Hausdorff distance(HD),mean intersection over union(MIoU),and mean pixel accuracy(MPA)values of 80.03%,6.649 mm,84.38%,and 89.41%,respectively.In terms of classification performance,our method achieves accuracy,recall,precision,and F1-score of 89.87%,90.44%,88.56%,and 89.49%,respectively.Compared with other networks,MMRAN performs better in segmentation and classification,which significantly aids medical professionals in brain tumor management.The code and data set are available at https://github.com/linkenfaqiu/MMRAN.展开更多
基金supported by the Outstanding Youth Science and Technology Innovation Team Project of Colleges and Universities in Hubei Province(Grant no.T201923)Key Science and Technology Project of Jingmen(Grant nos.2021ZDYF024,2022ZDYF019)Cultivation Project of Jingchu University of Technology(Grant no.PY201904).
文摘Brain tumors,one of the most lethal diseases with low survival rates,require early detection and accurate diagnosis to enable effective treatment planning.While deep learning architectures,particularly Convolutional Neural Networks(CNNs),have shown significant performance improvements over traditional methods,they struggle to capture the subtle pathological variations between different brain tumor types.Recent attention-based models have attempted to address this by focusing on global features,but they come with high computational costs.To address these challenges,this paper introduces a novel parallel architecture,ParMamba,which uniquely integrates Convolutional Attention Patch Embedding(CAPE)and the Conv Mamba block including CNN,Mamba and the channel enhancement module,marking a significant advancement in the field.The unique design of ConvMamba block enhances the ability of model to capture both local features and long-range dependencies,improving the detection of subtle differences between tumor types.The channel enhancement module refines feature interactions across channels.Additionally,CAPE is employed as a downsampling layer that extracts both local and global features,further improving classification accuracy.Experimental results on two publicly available brain tumor datasets demonstrate that ParMamba achieves classification accuracies of 99.62%and 99.35%,outperforming existing methods.Notably,ParMamba surpasses vision transformers(ViT)by 1.37%in accuracy,with a throughput improvement of over 30%.These results demonstrate that ParMamba delivers superior performance while operating faster than traditional attention-based methods.
基金the Deputyship for Research&Innovation,Ministry of Education in Saudi Arabia for funding this research work through the project Number PNU-DRI-RI-20-029.
文摘Detection of brain tumors in MRI images is the first step in brain cancer diagnosis.The accuracy of the diagnosis depends highly on the expertise of radiologists.Therefore,automated diagnosis of brain cancer from MRI is receiving a large amount of attention.Also,MRI tumor detection is usually followed by a biopsy(an invasive procedure),which is a medical procedure for brain tumor classification.It is of high importance to devise automated methods to aid radiologists in brain cancer tumor diagnosis without resorting to invasive procedures.Convolutional neural network(CNN)is deemed to be one of the best machine learning algorithms to achieve high-accuracy results in tumor identification and classification.In this paper,a CNN-based technique for brain tumor classification has been developed.The proposed CNN can distinguish between normal(no-cancer),astrocytoma tumors,gliomatosis cerebri tumors,and glioblastoma tumors.The implemented CNN was tested on MRI images that underwent a motion-correction procedure.The CNN was evaluated using two performance measurement procedures.The first one is a k-fold cross-validation testing method,in which we tested the dataset using k=8,10,12,and 14.The best accuracy for this procedure was 96.26%when k=10.To overcome the over-fitting problem that could be occurred in the k-fold testing method,we used a hold-out testing method as a second evaluation procedure.The results of this procedure succeeded in attaining 97.8%accuracy,with a specificity of 99.2%and a sensitivity of 97.32%.With this high accuracy,the developed CNN architecture could be considered an effective automated diagnosis method for the classification of brain tumors from MRI images.
文摘Knowledge-based transfer learning techniques have shown good performance for brain tumor classification,especially with small datasets.However,to obtain an optimized model for targeted brain tumor classification,it is challenging to select a pre-trained deep learning(DL)model,optimal values of hyperparameters,and optimization algorithm(solver).This paper first presents a brief review of recent literature related to brain tumor classification.Secondly,a robust framework for implementing the transfer learning technique is proposed.In the proposed framework,a Cartesian product matrix is generated to determine the optimal values of the two important hyperparameters:batch size and learning rate.An extensive exercise consisting of 435 simulations for 11 state-of-the-art pre-trained DL models was performed using 16 paired hyperparameters from the Cartesian product matrix to input the model with the three most popular solvers(stochastic gradient descent with momentum(SGDM),adaptive moment estimation(ADAM),and root mean squared propagation(RMSProp)).The 16 pairs were formed using individual hyperparameter values taken from literature,which generally addressed only one hyperparameter for optimization,rather than making a grid for a particular range.The proposed framework was assessed using a multi-class publicly available dataset consisting of glioma,meningioma,and pituitary tumors.Performance assessment shows that ResNet18 outperforms all other models in terms of accuracy,precision,specificity,and recall(sensitivity).The results are also compared with existing state-of-the-art research work that used the same dataset.The comparison was mainly based on performance metric“accuracy”with support of three other parameters“precision,”“recall,”and“specificity.”The comparison shows that the transfer learning technique,implemented through our proposed framework for brain tumor classification,outperformed all existing approaches.To the best of our knowledge,the proposed framework is an efficient framework that helped reduce the computational complexity and the time to attain optimal values of two important hyperparameters and consequently the optimized model with an accuracy of 99.56%.
文摘The deep learning models are identified as having a significant impact on various problems.The same can be adapted to the problem of brain tumor classification.However,several deep learning models are presented earlier,but they need better classification accuracy.An efficient Multi-Feature Approximation Based Convolution Neural Network(CNN)model(MFACNN)is proposed to handle this issue.The method reads the input 3D Magnetic Resonance Imaging(MRI)images and applies Gabor filters at multiple levels.The noise-removed image has been equalized for its quality by using histogram equalization.Further,the features like white mass,grey mass,texture,and shape are extracted from the images.Extracted features are trained with deep learning Convolution Neural Network(CNN).The network has been designed with a single convolution layer towards dimensionality reduction.The texture features obtained from the brain image have been transformed into a multi-dimensional feature matrix,which has been transformed into a single-dimensional feature vector at the convolution layer.The neurons of the intermediate layer are designed to measure White Mass Texture Support(WMTS),GrayMass Texture Support(GMTS),WhiteMass Covariance Support(WMCS),GrayMass Covariance Support(GMCS),and Class Texture Adhesive Support(CTAS).In the test phase,the neurons at the intermediate layer compute the support as mentioned above values towards various classes of images.Based on that,the method adds a Multi-Variate Feature Similarity Measure(MVFSM).Based on the importance ofMVFSM,the process finds the class of brain image given and produces an efficient result.
文摘Deep learning(DL)techniques,which do not need complex preprocessing and feature analysis,are used in many areas of medicine and achieve promising results.On the other hand,in medical studies,a limited dataset decreases the abstraction ability of the DL model.In this context,we aimed to produce synthetic brain images including three tumor types(glioma,meningioma,and pituitary),unlike traditional data augmentation methods,and classify them with DL.This study proposes a tumor classification model consisting of a Dense Convolutional Network(DenseNet121)-based DL model to prevent forgetting problems in deep networks and delay information flow between layers.By comparing models trained on two different datasets,we demonstrated the effect of synthetic images generated by Cycle Generative Adversarial Network(CycleGAN)on the generalization of DL.One model is trained only on the original dataset,while the other is trained on the combined dataset of synthetic and original images.Synthetic data generated by CycleGAN improved the best accuracy values for glioma,meningioma,and pituitary tumor classes from 0.9633,0.9569,and 0.9904 to 0.9968,0.9920,and 0.9952,respectively.The developed model using synthetic data obtained a higher accuracy value than the related studies in the literature.Additionally,except for pixel-level and affine transform data augmentation,synthetic data has been generated in the figshare brain dataset for the first time.
文摘Background Brain tumors are challenging to diagnose and treat,and require accurate and early therapeutic intervention.Magnetic Resonance Imaging(MRI)scans can visualize the internal structure of the brain.Often,deep learning is applied to images for the early and accurate detection of tumor cells.However,these models lack accuracy and efficacy in practical applications.Hybrid or modified models can facilitate better classification and provide insights into early-stage cancer detection.Methods This study demonstrates a parallel architecture that uses MRI images and integrates transformer-based frameworks with Convolutional Neural Networks(CNNs)to better classify distinct types of brain tumors.The proposed architecture,SwinResDual(SwRD),combines a Residual Network(ResNet)and a Swin Transformer in parallel to extract key features from input images.Using augmented MRI scans,31,464 scans for multiclass classification,and 30,000 scans for binary classification,the architecture simultaneously processed images through the ResNet50 and Swin Transformer branches,leveraging their strengths in hierarchical feature extraction and global context modeling to efficiently capture local and global image features.The final classifications are obtained by merging these features and passing them through a linear classifier.This approach identifies strong and varied characteristics and provides a precise brain tumor diagnosis.Results In the extensive evaluation,the model performed with an accuracy of 99.79% and a cross-validation accuracy of 100%for multiclass classification,along with 99.97% accuracy in binary classification.Conclusions In conclusion,the findings demonstrate great promise for brain tumor detection and advanced medical imaging diagnostics.
文摘Brain tumor classification is crucial for personalized treatment planning.Although deep learning-based Artificial Intelligence(AI)models can automatically analyze tumor images,fine details of small tumor regions may be overlooked during global feature extraction.Therefore,we propose a brain tumor Magnetic Resonance Imaging(MRI)classification model based on a global-local parallel dual-branch structure.The global branch employs ResNet50 with a Multi-Head Self-Attention(MHSA)to capture global contextual information from whole brain images,while the local branch utilizes VGG16 to extract fine-grained features from segmented brain tumor regions.The features from both branches are processed through designed attention-enhanced feature fusion module to filter and integrate important features.Additionally,to address sample imbalance in the dataset,we introduce a category attention block to improve the recognition of minority classes.Experimental results indicate that our method achieved a classification accuracy of 98.04%and a micro-average Area Under the Curve(AUC)of 0.989 in the classification of three types of brain tumors,surpassing several existing pre-trained Convolutional Neural Network(CNN)models.Additionally,feature interpretability analysis validated the effectiveness of the proposed model.This suggests that the method holds significant potential for brain tumor image classification.
基金funded through Researchers Supporting Project Number(RSPD2024R996)King Saud University,Riyadh,Saudi Arabia。
文摘Breast cancer detection heavily relies on medical imaging, particularly ultrasound, for early diagnosis and effectivetreatment. This research addresses the challenges associated with computer-aided diagnosis (CAD) of breastcancer fromultrasound images. The primary challenge is accurately distinguishing between malignant and benigntumors, complicated by factors such as speckle noise, variable image quality, and the need for precise segmentationand classification. The main objective of the research paper is to develop an advanced methodology for breastultrasound image classification, focusing on speckle noise reduction, precise segmentation, feature extraction, andmachine learning-based classification. A unique approach is introduced that combines Enhanced Speckle ReducedAnisotropic Diffusion (SRAD) filters for speckle noise reduction, U-NET-based segmentation, Genetic Algorithm(GA)-based feature selection, and Random Forest and Bagging Tree classifiers, resulting in a novel and efficientmodel. To test and validate the hybrid model, rigorous experimentations were performed and results state thatthe proposed hybrid model achieved accuracy rate of 99.9%, outperforming other existing techniques, and alsosignificantly reducing computational time. This enhanced accuracy, along with improved sensitivity and specificity,makes the proposed hybrid model a valuable addition to CAD systems in breast cancer diagnosis, ultimatelyenhancing diagnostic accuracy in clinical applications.
基金the“Intelligent Recognition Industry Service Center”as part of the Featured Areas Research Center Program under the Higher Education Sprout Project by the Ministry of Education(MOE)in Taiwan,and the National Science and Technology Council,Taiwan,under grants 113-2221-E-224-041 and 113-2622-E-224-002.Additionally,partial support was provided by Isuzu Optics Corporation.
文摘Liver cancer remains a leading cause of mortality worldwide,and precise diagnostic tools are essential for effective treatment planning.Liver Tumors(LTs)vary significantly in size,shape,and location,and can present with tissues of similar intensities,making automatically segmenting and classifying LTs from abdominal tomography images crucial and challenging.This review examines recent advancements in Liver Segmentation(LS)and Tumor Segmentation(TS)algorithms,highlighting their strengths and limitations regarding precision,automation,and resilience.Performance metrics are utilized to assess key detection algorithms and analytical methods,emphasizing their effectiveness and relevance in clinical contexts.The review also addresses ongoing challenges in liver tumor segmentation and identification,such as managing high variability in patient data and ensuring robustness across different imaging conditions.It suggests directions for future research,with insights into technological advancements that can enhance surgical planning and diagnostic accuracy by comparing popular methods.This paper contributes to a comprehensive understanding of current liver tumor detection techniques,provides a roadmap for future innovations,and improves diagnostic and therapeutic outcomes for liver cancer by integrating recent progress with remaining challenges.
基金supported by the Sanming Project of Medicine in Shenzhen (SZSM202211017)。
文摘Accurate cancer staging is the foundation of precision oncology and guides prognosis prediction and therapeutic decision-making. The conjoint TNM System by the American Joint Committee on Cancer (AJCC) and the International Union Against Cancer (UICC) has served as the global standard for tumor classification since inception.
文摘The latest edition of the WHO classification of the central nervous system was published in 2021.This review summarizes the major revisions to the classification of anterior pituitary tumors.The most important revision involves preferring the terminology of pituitary neuroendocrine tumor(PitNET),even though the terminology of pituitary adenoma(PA)still can be used according to this WHO classification compared to the previous one.Moreover,immunohistochemistry(IHC)examination of pituitary-specific transcription factors(TFs),including PIT1,TPIT,SF-1,GATA2/3,and ERα,is endorsed to determine the tumor cell lineage and to facilitate the classification of PitNET/PA subgroups.However,TF-negative IHC staining indicates PitNET/PA with no distinct cell lineages,which includes unclassified plurihormonal(PH)tumors and null cell(NC)tumors in this edition.The new WHO classification of PitNET/PA has incorporated tremendous advances in the understanding of the cytogenesis and pathogenesis of pituitary tumors.However,due to the shortcomings of the technology used in the diagnosis of PitNET/PA and the limited understanding of the tumorigenesis of PitNET/PA,the application of this new classification system in practice should be further evaluated and validated.Besides providing information for deciding the follow-up plans and adjunctive treatment after surgery,this classification system offers no additional help for neurosurgeons in clinical practice,especially in determining the treatment strategies.Therefore,it is necessary for neurosurgeons to establish a comprehensive pituitary classification system for PitNET/PA that incorporates neuroimaging grading data or direct observation of invasiveness during operation or the predictor of prognosis,as well as pathological diagnosis,thereby distinguishing the invasiveness of the tumor and facilitating neurosurgeons to decide on the treatment strategies and follow-up plans as well as adjunctive treatment after surgery.
基金This work was supported in part by the National Natural Science Foundation of China(No.11701144)National Science Foundation of US(No.DMS1719932)+1 种基金Natural Science Foundation of Henan Province(No.162300410061)Project of Emerging Interdisciplinary(No.xxjc20170003).
文摘Image-based breast tumor classification is an active and challenging problem.In this paper,a robust breast tumor classification framework is presented based on deep feature representation learning and exploiting available information in existing samples.Feature representation learning of mammograms is fulfilled by a modified nonnegative matrix factorization model called LPML-LRNMF,which is motivated by hierarchical learning and layer-wise pre-training(LP)strategy in deep learning.Low-rank(LR)constraint is integrated into the feature representation learning model by considering the intrinsic characteristics of mammograms.Moreover,the proposed LPML-LRNMF model is optimized via alternating direction method of multipliers and the corresponding convergence is analyzed.For completing classification,an inverse projection sparse representation model is introduced to exploit information embedded in existing samples,especially in test ones.Experiments on the public dataset and actual clinical dataset show that the classification accuracy,specificity and sensitivity achieve the clinical acceptance level.
文摘Objective To explore classification and surgical approach of magnum foramen tumor. Methods A retrospective analysis was performed for 43 surgically treated patients with tumors involving foramen magnum. According to the site of tumor,the classification was divided into:Type Ⅰ,located at dorsal,Ⅰ a extra-medullary,
文摘This study investigated the accuracy of MRI features in differentiating the pathological grades of pancreatic neuroendocrine neoplasms（PNENs）. A total of 31 PNENs patients were retrospectively evaluated, including 19 cases in grade 1, 5 in grade 2, and 7 in grade 3. Plain and contrastenhanced MRI was performed on all patients. MRI features including tumor size, margin, signal intensity, enhancement patterns, degenerative changes, duct dilatation and metastasis were analyzed. Chi square tests, Fisher＇s exact tests, one-way ANOVA and ROC analysis were conducted to assess the associations between MRI features and different tumor grades. It was found that patients with older age, tumors with higher TNM stage and without hormonal syndrome had higher grade of PNETs（all P〈0.05）. Tumor size, shape, margin and growth pattern, tumor pattern, pancreatic and bile duct dilatation and presence of lymphatic and distant metastasis as well as MR enhancement pattern and tumor-topancreas contrast during arterial phase were the key features differentiating tumors of all grades（all P〈0.05）. ROC analysis revealed that the tumor size with threshold of 2.8 cm, irregular shape, pancreatic duct dilatation and lymphadenopathy showed satisfactory sensitivity and specificity in distinguishing grade 3 from grade 1 and grade 2 tumors. Features of peripancreatic tissue or vascular invasion, and distant metastasis showed high specificity but relatively low sensitivity. In conclusion, larger size, poorlydefined margin, heterogeneous enhanced pattern during arterial phase, duct dilatation and the presence of metastases are common features of higher grade PNENs. Plain and contrast-enhanced MRI provides the ability to differentiate tumors with different pathological grades.
文摘Gene expression microarray data can be used to classify tumor types. We proposed a new procedure to classify human tumor samples based on microarray gene expressions by using a hybrid supervised learning method called MOEA+WV (Multi-Objective Evolutionary Algorithm+Weighted Voting). MOEA is used to search for a relatively few subsets of informative genes from the high-dimensional gene space, and WV is used as a classification tool. This new method has been applied to predicate the subtypes of lymphoma and outcomes of medulloblastoma. The results are relatively accurate and meaningful compared to those from other methods. Key words bioinformatics - tumor classification - Pareto optimization - MOEA CLC number Q 786 - TP 181 Foundation item: Supported by the National Natural Science Foundation of China (60301009), the Foundation of Young Scholars of Ministry of Education of China (150118) and Chenguang Project of Wuhan City (211121009).Biography: Liu Juan (1970-), female, Associate Professor, Postdoctoral, research direction: bioinformatics, data mining, machine learning.
文摘This systematic review aims to comprehensively examine and compare deep learning methods for brain tumor segmentation and classification using MRI and other imaging modalities,focusing on recent trends from 2022 to 2025.The primary objective is to evaluate methodological advancements,model performance,dataset usage,and existing challenges in developing clinically robust AI systems.We included peer-reviewed journal articles and highimpact conference papers published between 2022 and 2025,written in English,that proposed or evaluated deep learning methods for brain tumor segmentation and/or classification.Excluded were non-open-access publications,books,and non-English articles.A structured search was conducted across Scopus,Google Scholar,Wiley,and Taylor&Francis,with the last search performed in August 2025.Risk of bias was not formally quantified but considered during full-text screening based on dataset diversity,validation methods,and availability of performance metrics.We used narrative synthesis and tabular benchmarking to compare performance metrics(e.g.,accuracy,Dice score)across model types(CNN,Transformer,Hybrid),imaging modalities,and datasets.A total of 49 studies were included(43 journal articles and 6 conference papers).These studies spanned over 9 public datasets(e.g.,BraTS,Figshare,REMBRANDT,MOLAB)and utilized a range of imaging modalities,predominantly MRI.Hybrid models,especially ResViT and UNetFormer,consistently achieved high performance,with classification accuracy exceeding 98%and segmentation Dice scores above 0.90 across multiple studies.Transformers and hybrid architectures showed increasing adoption post2023.Many studies lacked external validation and were evaluated only on a few benchmark datasets,raising concerns about generalizability and dataset bias.Few studies addressed clinical interpretability or uncertainty quantification.Despite promising results,particularly for hybrid deep learning models,widespread clinical adoption remains limited due to lack of validation,interpretability concerns,and real-world deployment barriers.
文摘One of the leading causes of mortality worldwide is liver cancer.The earlier the detection of hepatic tumors,the lower the mortality rate.This paper introduces a computer-aided diagnosis system to extract hepatic tumors from computed tomography scans and classify them into malignant or benign tumors.Segmenting hepatic tumors from computed tomography scans is considered a challenging task due to the fuzziness in the liver pixel range,intensity values overlap between the liver and neighboring organs,high noise from computed tomography scanner,and large variance in tumors shapes.The proposed method consists of three main stages;liver segmentation using Fast Generalized Fuzzy C-Means,tumor segmentation using dynamic thresholding,and the tumor’s classification into malignant/benign using support vector machines classifier.The performance of the proposed system was evaluated using three liver benchmark datasets,which are MICCAI-Sliver07,LiTS17,and 3Dircadb.The proposed computer adided diagnosis system achieved an average accuracy of 96.75%,sensetivity of 96.38%,specificity of 95.20%and Dice similarity coefficient of 95.13%.
基金supported by the following funds:the Key Research and Development Project of the Science and Technology Department of Sichuan Province(Grant Nos.2021YFS0202 and 2021YFS0229)the Natural Science Foundation of Sichuan Province(Grant No.2022NSFSC1326)+1 种基金Postdoctoral Research Fund of West China Hospital(Grant Nos.2019HXBH056 and 2020HXBH066)China Postdoctoral Science Foundation(Grant No.2022T150454).
文摘Lung adenocarcinoma (LUAD) is the most common and deadliest subtype of lung cancer. To select moretargeted and effective treatments for individuals, further advances in classifying LUAD are urgently needed. Thenumber, type, and function of T cells in the tumor microenvironment (TME) determine the progression andtreatment response of LUAD. Long noncoding RNAs (lncRNAs), may regulate T cell differentiation, development,and activation. Thus, our aim was to identify T cell-related lncRNAs (T cell-Lncs) in LUAD and to investigatewhether T cell-Lncs could serve as potential stratifiers and therapeutic targets. Seven T cell-Lncs were identified tofurther establish the T cell-related lncRNA risk score (TRS) in LUAD. Low TRS individuals were characterized byrobust immune status, fewer genomic alterations, and remarkably longer survival than high TRS individuals. Theexcellent accuracy of TRS in predicting overall survival (OS) was validated in the TCGA-LUAD training cohort andthe GEO-LUAD validation cohort. Our data demonstrated the favorable predictive power of the TRS-basednomogram, which had important clinical significance in estimating the survival probability for individuals. Inaddition, individuals with low TRS could respond better to chemotherapy and immunotherapy than those with highTRS. LINC00525 was identified as a valuable study target, and the ability of LUAD to proliferate or invade wassignificantly attenuated by downregulation of LINC00525. In conclusion, the TRS established by T cell-Lncs couldunambiguously classify LUAD patients, predict their prognosis and guide their management. Moreover, our identifiedT cell-Lncs could provide potential therapeutic targets for LUAD.
基金The research performed in this lab is supported by Shanghai Science Foundation(NO.04DZ14006)National Natural Science Foundation(NO.30450001)+1 种基金Major State Basic Research Development program of China(NO.2004CB51 8804)the National High Technology Re-search and Development Program of China(NO.2002AA2Z3352).
文摘DNA methylation is the most intensively studied epigenetic phenomenon, disturbances of which result in changes ingene transcription, thus exerting drastic imparts onto biological behaviors of cancer. Both the global demethylation andthe local hypermethylation have been widely reported in all types of tumors, providing both challenges and opportunitiesfor a better understanding and eventually controlling of the malignance. However, we are still in the very early stage ofinformation accumulation concerning the tumor associated changes in DNA methylation pattern. A number of excellentrecent reviews have covered this issue in depth. Therefore, this review will summarize our recent data on DNA methy-lation profiling in cancers. Perspectives for the future direction in this dynamic and exciting field will also be given.
基金This paper was supported by National Natural Science Foundation of China(No.61977063 and 61872020).The authors thank all the patients for providing their MRI images and School of Biomedical Engineering at Southern Medical University,China for providing the brain tumor data set.We appreciate Dr.Fenfen Li,Wenzhou Eye Hospital,Wenzhou Medical University,China,for her support with clinical consulting and language editing.
文摘Automatic segmentation and classification of brain tumors are of great importance to clinical treatment.However,they are challenging due to the varied and small morphology of the tumors.In this paper,we propose a multitask multiscale residual attention network(MMRAN)to simultaneously solve the problem of accurately segmenting and classifying brain tumors.The proposed MMRAN is based on U-Net,and a parallel branch is added at the end of the encoder as the classification network.First,we propose a novel multiscale residual attention module(MRAM)that can aggregate contextual features and combine channel attention and spatial attention better and add it to the shared parameter layer of MMRAN.Second,we propose a method of dynamic weight training that can improve model performance while minimizing the need for multiple experiments to determine the optimal weights for each task.Finally,prior knowledge of brain tumors is added to the postprocessing of segmented images to further improve the segmentation accuracy.We evaluated MMRAN on a brain tumor data set containing meningioma,glioma,and pituitary tumors.In terms of segmentation performance,our method achieves Dice,Hausdorff distance(HD),mean intersection over union(MIoU),and mean pixel accuracy(MPA)values of 80.03%,6.649 mm,84.38%,and 89.41%,respectively.In terms of classification performance,our method achieves accuracy,recall,precision,and F1-score of 89.87%,90.44%,88.56%,and 89.49%,respectively.Compared with other networks,MMRAN performs better in segmentation and classification,which significantly aids medical professionals in brain tumor management.The code and data set are available at https://github.com/linkenfaqiu/MMRAN.
