Background Accurately identifying and localizing the five subtypes of intracranial hemorrhage(ICH)are crucial steps for subsequent clinical treatment;however,the lack of a large computed tomography(CT)dataset with ann...Background Accurately identifying and localizing the five subtypes of intracranial hemorrhage(ICH)are crucial steps for subsequent clinical treatment;however,the lack of a large computed tomography(CT)dataset with annotations of the categorization and localization of ICH considerably limits the development of deep learningbased identification and localization methods.We aimed to construct this large dataset and develop a deep learning-based model to identify and localize the five ICH subtypes,including intraventricular hemorrhage(IVH),intraparenchymal hemorrhage(IPH),subdural hemorrhage(SDH),subarachnoid hemorrhage(SAH),and epidural hemorrhage(EDH),in non-contrast head CT scans.Methods Based on the public Radiological Society of North America(RSNA)2019 dataset,we constructed a large CT dataset named RSNA 2019+that was annotated for bleeding localization of the five ICH subtypes by three radiologists.An improved YOLOv8 architecture with the bidirectional feature pyramid network was proposed and trained using the RSNA 2019+training dataset and evaluated on the RSNA 2019+test dataset.The public CQ500,and two private datasets collected from the Xinqiao and Sunshine Union Hospitals,respectively,were also annotated to perform multicenter validation.Furthermore,the performance of the deep learning model was compared with that of four radiologists.Multiple performance metrics,including the average precision(AP),precision,recall and F1-score,were used for performance evaluation.The McNemar and chi-squared tests were performed,and the 95%Wilson confidence intervals were given for the precision and recall.Results There were 175,125;4,707;8,259;and 3,104 bounding boxes after annotation on the RSNA 2019+;CQ500+;and the PD 1 and PD 2 datasets,respectively.With an intersection-over-union threshold of 0.5,the APs of IVH,IPH,SAH,SDH and EDH are 0.852,0.820,0.574,0.639,and 0.558,respectively,yielding a mean average precision(mAP)of 0.688 for our proposed deep learning model on the RSNA 2019+test dataset.For the multicenter validation involving the three external datasets,the mAPs for CQ500,PD1,and PD2 were 0.594,0.734,and 0.66,respectively,which is comparable to those of radiologist with eight years of experience in head CT interpretation.Conclusion The deep learning model developed from the constructed RSNA 2019+dataset exhibited good potential in identifying and localizing the five ICH subtypes in CT slices and has the potential to assist in the clinical diagnosis.展开更多
Malformed fruits depreciate a plant’s market value.In tomato(Solanum lycopersicum),fruit malformation is associated with the multi-locule trait,which involves genes regulating shoot apical meristem(SAM)development.Th...Malformed fruits depreciate a plant’s market value.In tomato(Solanum lycopersicum),fruit malformation is associated with the multi-locule trait,which involves genes regulating shoot apical meristem(SAM)development.The expression pattern of TOPLESS3(SITPL3)throughout SAM development prompted us to investigate its functional significance via RNA interference(RNAi)and clustered regularly interspaced short palindromic repeats/CRISPR-associated nuclease 9(Cas9)-mediated gene editing.Lower SITPL3 transcript levels resulted in larger fruits with more locules and larger SAMs at the 5 d after germination(DAG5)stage.Differentially expressed genes in the SAM of wild-type(WT)and SITPL3-RNAi plants,identified by transcriptome deep sequencing(RNA-seq),were enriched in the gibberellin(GA)biosynthesis and plant hormone signaling pathways.Moreover,exogenous auxin and paclobutrazol treatments rescued the multi-locule phenotype,indicating that SITPL3 affects SAM size by mediating auxin and GA levels in the SAM.Furthermore,SITPL3 interacted with WUSCHEL(SIWUS),which plays an important role in SAM size maintenance.We conducted RNA-seq and DNA affinity purification followed by sequencing(DAP-seq)analyses to identify the genes regulated by SITPL3 and SIWUS in the SAM and to determine how they regulate SAM size.We detected24 overlapping genes regulated by SITPL3 and SIWUS and harboring an SIWUS-binding motif in their promoters.Furthermore,functional annotation revealed a notable enrichment for functions in auxin transport,auxin signal transduction,and GA biosynthesis.Dual-luciferase assays also revealed that SITPL3 enhances SIWUS-mediated regulation(repression and activation)of SIPIN3 and SIGA2 ox4 transcription,indicating that the SITPL3-SIWUS module regulates SAM size by mediating auxin distribution and GA levels,and perturbations of this module result in enlarged SAM.These results provide novel insights into the molecular mechanism of SAM maintenance and locule formation in tomato and highlight the SITPL3-SIWUS module as a key regulator.展开更多
The prominent third-order nonlinear optical properties of WTe2 films are studied through the Z-scan technique using a femtosecond pulsed laser at 1030 nm.Open-aperture(OA)and closed-aperture(CA)Z-scan measurements are...The prominent third-order nonlinear optical properties of WTe2 films are studied through the Z-scan technique using a femtosecond pulsed laser at 1030 nm.Open-aperture(OA)and closed-aperture(CA)Z-scan measurements are performed at different intensities to investigate the nonlinear absorption and refraction properties of WTe2 films.OA Z-scan results show that WTe2 films always hold a saturable absorption characteristic without transition to reverse saturable absorption.Further,a large nonlinear absorption coefficientβis determined to be-3.37×10^3 cm∕GW by fitting the OA Z-scan curve at the peak intensity of 15.603 GW∕cm^2.In addition,through the slow saturation absorption model,the ground state absorption cross section,excited state absorption cross section,and absorber's density were found to be 1.4938×10^-16 cm^2,1.2536×10^-16 cm^2,and 6.2396×10^20 cm^-3,respectively.CA Z-scan results exhibit a classic peak–valley shape of the CA Z-scan signal,which reveals a self-defocusing optical effect of WTe2 films under the measured environment.Furthermore,a considerable nonlinear refractive index value n2 can be obtained at-1.629×10^-2 cm^2∕GW.Ultimately,the values of the real and imaginary parts of the third-order nonlinear susceptibility of WTe2 are roughly estimated to be-5.93×10^-9 and-1.01×10^-8 esu with the help of first-principles calculations.WTe2 possesses sizeable thirdorder nonlinear coefficients compared with other materials,which suggests that it has great potential to improve performance in nonlinear optical devices.展开更多
基金supported by Science and Technology Research Project of Chongqing Education Commission(Grant No.KJQN202212804).
文摘Background Accurately identifying and localizing the five subtypes of intracranial hemorrhage(ICH)are crucial steps for subsequent clinical treatment;however,the lack of a large computed tomography(CT)dataset with annotations of the categorization and localization of ICH considerably limits the development of deep learningbased identification and localization methods.We aimed to construct this large dataset and develop a deep learning-based model to identify and localize the five ICH subtypes,including intraventricular hemorrhage(IVH),intraparenchymal hemorrhage(IPH),subdural hemorrhage(SDH),subarachnoid hemorrhage(SAH),and epidural hemorrhage(EDH),in non-contrast head CT scans.Methods Based on the public Radiological Society of North America(RSNA)2019 dataset,we constructed a large CT dataset named RSNA 2019+that was annotated for bleeding localization of the five ICH subtypes by three radiologists.An improved YOLOv8 architecture with the bidirectional feature pyramid network was proposed and trained using the RSNA 2019+training dataset and evaluated on the RSNA 2019+test dataset.The public CQ500,and two private datasets collected from the Xinqiao and Sunshine Union Hospitals,respectively,were also annotated to perform multicenter validation.Furthermore,the performance of the deep learning model was compared with that of four radiologists.Multiple performance metrics,including the average precision(AP),precision,recall and F1-score,were used for performance evaluation.The McNemar and chi-squared tests were performed,and the 95%Wilson confidence intervals were given for the precision and recall.Results There were 175,125;4,707;8,259;and 3,104 bounding boxes after annotation on the RSNA 2019+;CQ500+;and the PD 1 and PD 2 datasets,respectively.With an intersection-over-union threshold of 0.5,the APs of IVH,IPH,SAH,SDH and EDH are 0.852,0.820,0.574,0.639,and 0.558,respectively,yielding a mean average precision(mAP)of 0.688 for our proposed deep learning model on the RSNA 2019+test dataset.For the multicenter validation involving the three external datasets,the mAPs for CQ500,PD1,and PD2 were 0.594,0.734,and 0.66,respectively,which is comparable to those of radiologist with eight years of experience in head CT interpretation.Conclusion The deep learning model developed from the constructed RSNA 2019+dataset exhibited good potential in identifying and localizing the five ICH subtypes in CT slices and has the potential to assist in the clinical diagnosis.
基金funded by the National Natural Science Foundation of China(31870286 and 31902013)the Natural Science Foundation of Guangdong Province(2022A1515012278,2017A030313114,2018A030310205,and 2021A1515010528)the General Project of Guangzhou City(201804010031)。
文摘Malformed fruits depreciate a plant’s market value.In tomato(Solanum lycopersicum),fruit malformation is associated with the multi-locule trait,which involves genes regulating shoot apical meristem(SAM)development.The expression pattern of TOPLESS3(SITPL3)throughout SAM development prompted us to investigate its functional significance via RNA interference(RNAi)and clustered regularly interspaced short palindromic repeats/CRISPR-associated nuclease 9(Cas9)-mediated gene editing.Lower SITPL3 transcript levels resulted in larger fruits with more locules and larger SAMs at the 5 d after germination(DAG5)stage.Differentially expressed genes in the SAM of wild-type(WT)and SITPL3-RNAi plants,identified by transcriptome deep sequencing(RNA-seq),were enriched in the gibberellin(GA)biosynthesis and plant hormone signaling pathways.Moreover,exogenous auxin and paclobutrazol treatments rescued the multi-locule phenotype,indicating that SITPL3 affects SAM size by mediating auxin and GA levels in the SAM.Furthermore,SITPL3 interacted with WUSCHEL(SIWUS),which plays an important role in SAM size maintenance.We conducted RNA-seq and DNA affinity purification followed by sequencing(DAP-seq)analyses to identify the genes regulated by SITPL3 and SIWUS in the SAM and to determine how they regulate SAM size.We detected24 overlapping genes regulated by SITPL3 and SIWUS and harboring an SIWUS-binding motif in their promoters.Furthermore,functional annotation revealed a notable enrichment for functions in auxin transport,auxin signal transduction,and GA biosynthesis.Dual-luciferase assays also revealed that SITPL3 enhances SIWUS-mediated regulation(repression and activation)of SIPIN3 and SIGA2 ox4 transcription,indicating that the SITPL3-SIWUS module regulates SAM size by mediating auxin distribution and GA levels,and perturbations of this module result in enlarged SAM.These results provide novel insights into the molecular mechanism of SAM maintenance and locule formation in tomato and highlight the SITPL3-SIWUS module as a key regulator.
基金International Cooperative Program(2014DFR10780)National Natural Science Foundation of China(61822403,U1732159)
文摘The prominent third-order nonlinear optical properties of WTe2 films are studied through the Z-scan technique using a femtosecond pulsed laser at 1030 nm.Open-aperture(OA)and closed-aperture(CA)Z-scan measurements are performed at different intensities to investigate the nonlinear absorption and refraction properties of WTe2 films.OA Z-scan results show that WTe2 films always hold a saturable absorption characteristic without transition to reverse saturable absorption.Further,a large nonlinear absorption coefficientβis determined to be-3.37×10^3 cm∕GW by fitting the OA Z-scan curve at the peak intensity of 15.603 GW∕cm^2.In addition,through the slow saturation absorption model,the ground state absorption cross section,excited state absorption cross section,and absorber's density were found to be 1.4938×10^-16 cm^2,1.2536×10^-16 cm^2,and 6.2396×10^20 cm^-3,respectively.CA Z-scan results exhibit a classic peak–valley shape of the CA Z-scan signal,which reveals a self-defocusing optical effect of WTe2 films under the measured environment.Furthermore,a considerable nonlinear refractive index value n2 can be obtained at-1.629×10^-2 cm^2∕GW.Ultimately,the values of the real and imaginary parts of the third-order nonlinear susceptibility of WTe2 are roughly estimated to be-5.93×10^-9 and-1.01×10^-8 esu with the help of first-principles calculations.WTe2 possesses sizeable thirdorder nonlinear coefficients compared with other materials,which suggests that it has great potential to improve performance in nonlinear optical devices.
