A novel variant c.902C>A (p. A301D) in KCNQ4 associated with non-syndromic deafness 2A in a Chinese family.

BACKGROUND: Deafness autosomal dominant 2A (DFNA2A) is related to non-syndromic genetic hearing impairment. The KCNQ4 (Potassium Voltage-Gated Channel Subfamily Q Member 4) can lead to DFNA2A. In this study, we report a case of autosomal dominant non-syndromic hearing loss with six family members as caused by a novel variant in the KCNQ4 gene. METHODS: The whole-exome sequencing (WES) and pure tone audiometry were performed on the proband of the family. Sanger sequencing was conducted on family members to determine if the novel variant in the KCNQ4 gene was present. Evolutionary conservation analysis and computational tertiary structure protein prediction of the wild-type KCNQ4 protein and its variant were then performed. In addition, voltage-gated channel activity of the wild-type KCNQ4 protein and its variant were tested using whole-cell patch clamp. RESULTS: It was observed that the proband had inherited autosomal dominant, non-syndromic sensorineural hearing loss as a trait. A novel co-segregating heterozygous missense variant (c.902C>A, p.Ala301Asp) of the KCNQ4 gene was identified in the proband and other five affected family members. This variant was predicted to cause an alanine-to-aspartic acid substitution at position 301 in the KCNQ4 protein. The alanine at position 301 is well conserved across different species. Whole-cell patch clamp showed that there was a significant difference between the WT protein currents and the mutant protein currents in the voltage-gated channel activity. CONCLUSION: In the present study, performing WES in conjunction with Sanger sequencing enhanced the detection of a novel, potentially causative variant (c301 A>G; p.Ala301Asp) in exon 6 of the KCNQ4 gene. Therefore, our findings contributed to the mutation spectrum of the KCNQ4 gene and may be useful in the diagnosis and gene therapy of deafness autosomal dominant 2A.

Identification of Novel Hb Guiyang [HBA2: c.151C > A α2 50 (CE8) His- Asn] and Phenotype- Genotype Correlation of Abnormal Hemoglobins in Guizhou, Southwest China.

Purpose: To analyze the composition of abnormal hemoglobin and the relationship between genotype and phenotype by screening abnormal hemoglobin in a subpopulation of Guizhou, China. Patients and Methods: Routine blood evaluation, capillary electrophoresis of hemoglobin, and mutation of α - and ß - thalassemia genes were evaluated in 19,976 individuals for thalassemia screening in Guizhou. Sanger sequencing of HBA1, HBA2 and HBB genes was performed in samples with abnormal bands or unexplained increases of normal bands. The types of abnormal hemoglobin were obtained by sequence analysis. Results: Abnormal hemoglobin was detected in 84 individuals (detection rate, 0.42%). Ten types each of α and ß globin chain variants were detected, including most commonly Hb E, Hb New York and Hb Port Phillip. In this study, the abnormal Hb Mizuho was identified for the first time in a Chinese population, and a novel abnormal hemoglobin Hb Guiyang (HBA2: c.151C > A) was detected for the first time. Except for Hb Mizuho, other abnormal hemoglobin heterozygotes without thalassemia or iron deficiency had no significant hematological changes. Conclusion: This study enriched the molecular epidemiological data of abnormal hemoglobin in Guizhou, China and provided reference data for genetic counseling and prenatal diagnosis of abnormal hemoglobin.

Long Non-Coding RNA H19 Leads to Upregulation of γ-Globin Gene Expression during Erythroid Differentiation.

Long noncoding RNAs (lncRNAs) are important because they are involved in a variety of life activities and have many downstream targets. Moreover, there is also increasing evidence that some lncRNAs play important roles in the expression and regulation of γ-globin genes. In our previous study, we analyzed genetic material from nucleated red blood cells (NRBCs) extracted from premature and full-term umbilical cord blood samples. Through RNA sequencing (RNA-Seq) analysis, lncRNA H19 emerged as a differentially expressed transcript between the two blood types. While this discovery provided insight into H19, previous studies had not investigated its effect on the γ-globin gene. Therefore, the focus of our study was to explore the impact of H19 on the γ-globin gene. In this study, we discovered that overexpressing H19 led to a decrease in HBG mRNA levels during erythroid differentiation in K562 cells. Conversely, in CD34+ hematopoietic stem cells and human umbilical cord blood-derived erythroid progenitor (HUDEP-2) cells, HBG expression increased. Additionally, we observed that H19 was primarily located in the nucleus of K562 cells, while in HUDEP-2 cells, H19 was present predominantly in the cytoplasm. These findings suggest a significant upregulation of HBG due to H19 overexpression. Notably, cytoplasmic localization in HUDEP-2 cells hints at its potential role as a competing endogenous RNA (ceRNA), regulating γ-globin expression by targeting microRNA/mRNA interactions.

A weakly supervised deep learning model integrating noncontrasted computed tomography images and clinical factors facilitates haemorrhagic transformation prediction after intravenous thrombolysis in acute ischaemic stroke patients.

BACKGROUND: Haemorrhage transformation (HT) is a serious complication of intravenous thrombolysis (IVT) in acute ischaemic stroke (AIS). Accurate and timely prediction of the risk of HT before IVT may change the treatment decision and improve clinical prognosis. We aimed to develop a deep learning method for predicting HT after IVT for AIS using noncontrast computed tomography (NCCT) images. METHODS: We retrospectively collected data from 828 AIS patients undergoing recombinant tissue plasminogen activator (rt-PA) treatment within a 4.5-h time window (n = 665) or of undergoing urokinase treatment within a 6-h time window (n = 163) and divided them into the HT group (n = 69) and non-HT group (n = 759). HT was defined based on the criteria of the European Cooperative Acute Stroke Study-II trial. To address the problems of indiscernible features and imbalanced data, a weakly supervised deep learning (WSDL) model for HT prediction was constructed based on multiple instance learning and active learning using admission NCCT images and clinical information in addition to conventional deep learning models. Threefold cross-validation and transfer learning were performed to confirm the robustness of the network. Of note, the predictive value of the commonly used scales in clinics associated with NCCT images (i.e., the HAT and SEDAN score) was also analysed and compared to measure the feasibility of our proposed DL algorithms. RESULTS: Compared to the conventional DL and ML models, the WSDL model had the highest AUC of 0.799 (95% CI 0.712-0.883). Significant differences were observed between the WSDL model and five ML models (P < 0.05). The prediction performance of the WSDL model outperforms the HAT and SEDAN scores at the optimal operating point (threshold = 1.5). Further subgroup analysis showed that the WSDL model performed better for symptomatic intracranial haemorrhage (AUC = 0.833, F1 score = 0.909). CONCLUSIONS: Our WSDL model based on NCCT images had relatively good performance for predicting HT in AIS and may be suitable for assisting in clinical treatment decision-making.

Identification of Two Novel Variants of the DMD Gene in Chinese Families with Duchenne Muscular Dystrophy.

Background: Duchenne muscular dystrophy (DMD), an X-linked recessive neuromuscular disorder, is caused by pathogenic variants in the DMD gene encoding a large structural protein in muscle cells. Methods: Two probands, a 6-year old boy and a 1-month old infant, respectively, were clinically diagnosed with DMD based on elevated levels of creatine kinase and creatine kinase isoenzyme. CNVplex and whole exome sequencing (WES) were performed for causal variants, and Sanger sequencing was used for verification. Results: CNVplex found no large deletions or duplications in the DMD gene in both patients, but WES discovered a single-nucleotide deletion in exon 48 (NM_004006.2:c.6963del, p.Asp2322ThrfsTer16) in the proband of pedigree 1, and a nonsense mutation in exon 27 (NM_004006.2:c.3637A>T, p.K1213Ter) in the proband of pedigree 2. Conclusion: The results of our study expand the mutation spectrum of DMD and enrich our understanding of the clinical characteristics of DMD. Genetic counseling was provided for the two families involved in this study.

A Rare Case of Abnormal Hemoglobin Variant Hb Mizuho: [HBB: c.206T > C ß 68(E12) Leu-Pro]: A First Report in the Chinese Population.

A 6-month-old female infant presented with unexplained hemolytic anemia, showing no abnormalities by capillary electrophoresis and genetic testing for α- and ß-thalassemia mutations that are commonly seen in the Chinese population. A rare Hb Mizuho: [HBB: c.206T > C ß 68(E12) Leu- Pro] variant was identified by next-generation sequencing (NGS) and verified by Sanger sequencing. Hb Mizuho: [HBB: c.206T > C ß 68(E12) Leu- Pro] is not easily detectable because it is extremely unstable, and the correct diagnosis is usually made via DNA sequencing. This is the first report of this variant in the Chinese population.

Deep learning-assisted LI-RADS grading and distinguishing hepatocellular carcinoma (HCC) from non-HCC based on multiphase CT: a two-center study.

OBJECTIVES: To develop a deep learning (DL) method that can determine the Liver Imaging Reporting and Data System (LI-RADS) grading of high-risk liver lesions and distinguish hepatocellular carcinoma (HCC) from non-HCC based on multiphase CT. METHODS: This retrospective study included 1049 patients with 1082 lesions from two independent hospitals that were pathologically confirmed as HCC or non-HCC. All patients underwent a four-phase CT imaging protocol. All lesions were graded (LR 4/5/M) by radiologists and divided into an internal (n = 886) and external cohort (n = 196) based on the examination date. In the internal cohort, Swin-Transformer based on different CT protocols were trained and tested for their ability to LI-RADS grading and distinguish HCC from non-HCC, and then validated in the external cohort. We further developed a combined model with the optimal protocol and clinical information for distinguishing HCC from non-HCC. RESULTS: In the test and external validation cohorts, the three-phase protocol without pre-contrast showed κ values of 0.6094 and 0.4845 for LI-RADS grading, and its accuracy was 0.8371 and 0.8061, while the accuracy of the radiologist was 0.8596 and 0.8622, respectively. The AUCs in distinguishing HCC from non-HCC were 0.865 and 0.715 in the test and external validation cohorts, while those of the combined model were 0.887 and 0.808. CONCLUSION: The Swin-Transformer based on three-phase CT protocol without pre-contrast could feasibly simplify LI-RADS grading and distinguish HCC from non-HCC. Furthermore, the DL model have the potential in accurately distinguishing HCC from non-HCC using imaging and highly characteristic clinical data as inputs. CLINICAL RELEVANCE STATEMENT: The application of deep learning model for multiphase CT has proven to improve the clinical applicability of the Liver Imaging Reporting and Data System and provide support to optimize the management of patients with liver diseases. KEY POINTS: • Deep learning (DL) simplifies LI-RADS grading and helps distinguish hepatocellular carcinoma (HCC) from non-HCC. • The Swin-Transformer based on the three-phase CT protocol without pre-contrast outperformed other CT protocols. • The Swin-Transformer provide help in distinguishing HCC from non-HCC by using CT and characteristic clinical information as inputs.

A novel heterozygous mutation of the NPHS1 gene in a Chinese child with congenital nephrotic syndrome: A case report.

RATIONALE: Congenital nephrotic syndrome (CNS) is a heterogeneous disorder in which massive proteinuria, hypoproteinemia, and hyperlipidemia and marked edema are the main manifestations before 3 months-of-age. Here, we present a case involving the genetic diagnosis of a child with CNS. PATIENT CONCERNS: A 31-day-old male infant with diarrhea for 25 days and generalized edema for more than 10 days. There was no family history of kidney disease. On proband whole exome sequencing, a compound heterozygous mutation of the NPHS1 gene was identified, including a novel in-frame mutation in exon 14 (c.1864_1866dupACC p. T622dup) and a missense mutation in exon 8 (c.928G>A p. D310N). DIAGNOSES: Based on the clinical and genetic findings, this patient was finally diagnosed with CNS. INTERVENTIONS: The main treatment options for the patient were 2-fold: anti-infective treatment and symptomatic treatment. OUTCOMES: The patient died in follow-up 2 months later; the specific reason for death was unclear. LESSONS: Whole exome sequencing and Sanger sequencing confirmed that the infant had CNS. Our study identified a novel mutation in an infant, thus expanding the gene-mutation spectrum of the NPHS1 gene, thus providing an efficient prenatal screening strategy and early genetic counseling.

Convolutional neural network for detecting rib fractures on chest radiographs: a feasibility study.

BACKGROUND: Chest radiography is the standard investigation for identifying rib fractures. The application of artificial intelligence (AI) for detecting rib fractures on chest radiographs is limited by image quality control and multilesion screening. To our knowledge, few studies have developed and verified the performance of an AI model for detecting rib fractures by using multi-center radiographs. And existing studies using chest radiographs for multiple rib fracture detection have used more complex and slower detection algorithms, so we aimed to create a multiple rib fracture detection model by using a convolutional neural network (CNN), based on multi-center and quality-normalised chest radiographs. METHODS: A total of 1080 radiographs with rib fractures were obtained and randomly divided into the training set (918 radiographs, 85%) and the testing set (162 radiographs, 15%). An object detection CNN, You Only Look Once v3 (YOLOv3), was adopted to build the detection model. Receiver operating characteristic (ROC) and free-response ROC (FROC) were used to evaluate the model's performance. A joint testing group of 162 radiographs with rib fractures and 233 radiographs without rib fractures was used as the internal testing set. Furthermore, an additional 201 radiographs, 121 with rib fractures and 80 without rib fractures, were independently validated to compare the CNN model performance with the diagnostic efficiency of radiologists. RESULTS: The sensitivity of the model in the training and testing sets was 92.0% and 91.1%, respectively, and the precision was 68.0% and 81.6%, respectively. FROC in the testing set showed that the sensitivity for whole-lesion detection reached 91.3% when the false-positive of each case was 0.56. In the joint testing group, the case-level accuracy, sensitivity, specificity, and area under the curve were 85.1%, 93.2%, 79.4%, and 0.92, respectively. At the fracture level and the case level in the independent validation set, the accuracy and sensitivity of the CNN model were always higher or close to radiologists' readings. CONCLUSIONS: The CNN model, based on YOLOv3, was sensitive for detecting rib fractures on chest radiographs and showed great potential in the preliminary screening of rib fractures, which indicated that CNN can help reduce missed diagnoses and relieve radiologists' workload. In this study, we developed and verified the performance of a novel CNN model for rib fracture detection by using radiography.

A Multimodal Multipath Artificial Intelligence System for Diagnosing Gastric Protruded Lesions on Endoscopy and Endoscopic Ultrasonography Images.

INTRODUCTION: The aim of this study was to develop a novel artificial intelligence (AI) system that can automatically detect and classify protruded gastric lesions and help address the challenges of diagnostic accuracy and inter-reader variability encountered in routine diagnostic workflow. METHODS: We analyzed data from 1,366 participants who underwent gastroscopy at Jiangsu Provincial People's Hospital and Yangzhou First People's Hospital between December 2010 and December 2020. These patients were diagnosed with submucosal tumors (SMTs) including gastric stromal tumors (GISTs), gastric leiomyomas (GILs), and gastric ectopic pancreas (GEP). We trained and validated a multimodal, multipath AI system (MMP-AI) using the data set. We assessed the diagnostic performance of the proposed AI system using the area under the receiver-operating characteristic curve (AUC) and compared its performance with that of endoscopists with more than 5 years of experience in endoscopic diagnosis. RESULTS: In the ternary classification task among subtypes of SMTs using modality images, MMP-AI achieved the highest AUCs of 0.896, 0.890, and 0.999 for classifying GIST, GIL, and GEP, respectively. The performance of the model was verified using both external and internal longitudinal data sets. Compared with endoscopists, MMP-AI achieved higher recognition accuracy for SMTs. DISCUSSION: We developed a system called MMP-AI to identify protruding benign gastric lesions. This system can be used not only for white-light endoscope image recognition but also for endoscopic ultrasonography image analysis.

Application of third-generation sequencing for genetic testing of thalassemia in Guizhou Province, Southwest China.

OBJECTIVES: To explore the application of third-generation sequencing (TGS) for genetic diagnosis and prenatal genetic screening of thalassemia genes. METHODS: Two groups of subjects were enrolled in this study. The first group included 176 subjects with positive hematological phenotypes for thalassemia. Thalassemia-associated genes were detected simultaneously in each sample using both the PacBio TGS platform based on single-molecule real-time (SMRT) technology and the conventional PCR-reverse dot blot (PCR-RDB). Sanger sequencing was used for validation when results were discordant between the two methods. The second group included 53 couples with at least one partner having a positive thalassemia hematological phenotype, and they were screened for homotypic thalassemia variants by TGS, and the risk of pregnancies with babies presenting with severe thalassemia, was assessed. RESULTS: Of the 176 subjects, 175 had concordant genotypes between the two methods, including 63 normal subjects and 112 α- and/or ß-thalassemia gene carriers, with a concordance rate of 99.43%. TGS detected a rare ß-thalassemia gene variant -50 (G > A) that was not detected by conventional PCR-RDB. TGS identified seven of the 53 couples as homotypic thalassemia gene carriers, five of whom were at risk of pregnancies with severe thalassemia. CONCLUSION: TGS could effectively detect common and rare thalassemia variants with high accuracy and efficiency. This approach would be suitable for prenatal thalassemia genetic screening in areas with high incidence of thalassemia.

Improving the diagnosis of acute ischemic stroke on non-contrast CT using deep learning: a multicenter study.

OBJECTIVE: This study aimed to develop a deep learning (DL) model to improve the diagnostic performance of EIC and ASPECTS in acute ischemic stroke (AIS). METHODS: Acute ischemic stroke patients were retrospectively enrolled from 5 hospitals. We proposed a deep learning model to simultaneously segment the infarct and estimate ASPECTS automatically using baseline CT. The model performance of segmentation and ASPECTS scoring was evaluated using dice similarity coefficient (DSC) and ROC, respectively. Four raters participated in the multi-reader and multicenter (MRMC) experiment to fulfill the region-based ASPECTS reading under the assistance of the model or not. At last, sensitivity, specificity, interpretation time and interrater agreement were used to evaluate the raters' reading performance. RESULTS: In total, 1391 patients were enrolled for model development and 85 patients for external validation with onset to CT scanning time of 176.4 ± 93.6 min and NIHSS of 5 (IQR 2-10). The model achieved a DSC of 0.600 and 0.762 and an AUC of 0.876 (CI 0.846-0.907) and 0.729 (CI 0.679-0.779), in the internal and external validation set, respectively. The assistance of the DL model improved the raters' average sensitivities and specificities from 0.254 (CI 0.22-0.26) and 0.896 (CI 0.884-0.907), to 0.333 (CI 0.301-0.345) and 0.915 (CI 0.904-0.926), respectively. The average interpretation time of the raters was reduced from 219.0 to 175.7 s (p = 0.035). Meanwhile, the interrater agreement increased from 0.741 to 0.980. CONCLUSIONS: With the assistance of our proposed DL model, radiologists got better performance in the detection of AIS lesions on NCCT.

Swin Transformer Improves the IDH Mutation Status Prediction of Gliomas Free of MRI-Based Tumor Segmentation.

Background: Deep learning (DL) could predict isocitrate dehydrogenase (IDH) mutation status from MRIs. Yet, previous work focused on CNNs with refined tumor segmentation. To bridge the gap, this study aimed to evaluate the feasibility of developing a Transformer-based network to predict the IDH mutation status free of refined tumor segmentation. Methods: A total of 493 glioma patients were recruited from two independent institutions for model development (TCIA; N = 259) and external test (AHXZ; N = 234). IDH mutation status was predicted directly from T2 images with a Swin Transformer and conventional ResNet. Furthermore, to investigate the necessity of refined tumor segmentation, seven strategies for the model input image were explored: (i) whole tumor slice; (ii-iii) tumor mask and/or not edema; (iv-vii) tumor bounding box of 0.8, 1.0, 1.2, 1.5 times. Performance comparison was made among the networks of different architectures along with different image input strategies, using area under the curve (AUC) and accuracy (ACC). Finally, to further boost the performance, a hybrid model was built by incorporating the images with clinical features. Results: With the seven proposed input strategies, seven Swin Transformer models and seven ResNet models were built, respectively. Based on the seven Swin Transformer models, an averaged AUC of 0.965 (internal test) and 0.842 (external test) were achieved, outperforming 0.922 and 0.805 resulting from the seven ResNet models, respectively. When a bounding box of 1.0 times was used, Swin Transformer (AUC = 0.868, ACC = 80.7%), achieved the best results against the one that used tumor segmentation (Tumor + Edema, AUC = 0.862, ACC = 78.5%). The hybrid model that integrated age and location features into images yielded improved performance (AUC = 0.878, Accuracy = 82.0%) over the model that used images only. Conclusions: Swin Transformer outperforms the CNN-based ResNet in IDH prediction. Using bounding box input images benefits the DL networks in IDH prediction and makes the IDH prediction free of refined glioma segmentation feasible.

Whole-exome sequencing identified five novel de novo variants in patients with unexplained intellectual disability.

BACKGROUND: Intellectual disability (ID) represents a neurodevelopmental disorder, which is characterized by marked defects in the intellectual function and adaptive behavior, with an onset during the developmental period. ID is mainly caused by genetic factors, and it is extremely genetically heterogeneous. This study aims to identify the genetic cause of ID using trio-WES analysis. METHODS: We recruited four pediatric patients with unexplained ID from non-consanguineous families, who presented at the Department of Pediatrics, Guizhou Provincial People's Hospital. Whole-exome sequencing (WES) and Sanger sequencing validation were performed in the patients and their unaffected parents. Furthermore, conservative analysis and protein structural and functional prediction were performed on the identified pathogenic variants. RESULTS: We identified five novel de novo mutations from four known ID-causing genes in the four included patients, namely COL4A1 (c.2786T>A, p.V929D and c.2797G>A, p.G933S), TBR1 (c.1639_1640insCCCGCAGTCC, p.Y553Sfs*124), CHD7 (c.7013A>T, p.Q2338L), and TUBA1A (c.1350del, p.E450Dfs*34). These mutations were all predicted to be deleterious and were located at highly conserved domains that might affect the structure and function of these proteins. CONCLUSION: Our findings contribute to expanding the mutational spectrum of ID-related genes and help to deepen the understanding of the genetic causes and heterogeneity of ID.

GMILT: A Novel Transformer Network That Can Noninvasively Predict EGFR Mutation Status.

Noninvasively and accurately predicting the epidermal growth factor receptor (EGFR) mutation status is a clinically vital problem. Moreover, further identifying the most suspicious area related to the EGFR mutation status can guide the biopsy to avoid false negatives. Deep learning methods based on computed tomography (CT) images may improve the noninvasive prediction of EGFR mutation status and potentially help clinicians guide biopsies by visual methods. Inspired by the potential inherent links between EGFR mutation status and invasiveness information, we hypothesized that the predictive performance of a deep learning network can be improved through extra utilization of the invasiveness information. Here, we created a novel explainable transformer network for EGFR classification named gated multiple instance learning transformer (GMILT) by integrating multi-instance learning and discriminative weakly supervised feature learning. Pathological invasiveness information was first introduced into the multitask model as embeddings. GMILT was trained and validated on a total of 512 patients with adenocarcinoma and tested on three datasets (the internal test dataset, the external test dataset, and The Cancer Imaging Archive (TCIA) public dataset). The performance (area under the curve (AUC) = 0.772 on the internal test dataset) of GMILT exceeded that of previously published methods and radiomics-based methods (i.e., random forest and support vector machine) and attained a preferable generalization ability (AUC = 0.856 in the TCIA test dataset and AUC = 0.756 in the external dataset). A diameter-based subgroup analysis further verified the efficiency of our model (most of the AUCs exceeded 0.772) to noninvasively predict EGFR mutation status from computed tomography (CT) images. In addition, because our method also identified the "core area" of the most suspicious area related to the EGFR mutation status, it has the potential ability to guide biopsies.

An original deep learning model using limited data for COVID-19 discrimination: A multicenter study.

OBJECTIVES: Artificial intelligence (AI) has been proved to be a highly efficient tool for COVID-19 diagnosis, but the large data size and heavy label force required for algorithm development and the poor generalizability of AI algorithms, to some extent, limit the application of AI technology in clinical practice. The aim of this study is to develop an AI algorithm with high robustness using limited chest CT data for COVID-19 discrimination. METHODS: A three dimensional algorithm that combined multi-instance learning with the LSTM architecture (3DMTM) was developed for differentiating COVID-19 from community acquired pneumonia (CAP) while logistic regression (LR), k-nearest neighbor (KNN), support vector machine (SVM), and a three dimensional convolutional neural network set for comparison. Totally, 515 patients with or without COVID-19 between December 2019 and March 2020 from five different hospitals were recruited and divided into relatively large (150 COVID-19 and 183 CAP cases) and relatively small datasets (17 COVID-19 and 35 CAP cases) for either training or validation and another independent dataset (37 COVID-19 and 93 CAP cases) for external test. Area under the receiver operating characteristic curve (AUC), sensitivity, specificity, precision, accuracy, F1 score, and G-mean were utilized for performance evaluation. RESULTS: In the external test cohort, the relatively large data-based 3DMTM-LD achieved an AUC of 0.956 (95% confidence interval, 95% CI, 0.929∼0.982) with 86.2% and 98.0% for its sensitivity and specificity. 3DMTM-SD got an AUC of 0.937 (95% CI, 0.909∼0.965), while the AUC of 3DCM-SD decreased dramatically to 0.714 (95% CI, 0.649∼0.780) with training data reduction. KNN-MMSD, LR-MMSD, SVM-MMSD, and 3DCM-MMSD benefited significantly from the inclusion of clinical information while models trained with relatively large dataset got slight performance improvement in COVID-19 discrimination. 3DMTM, trained with either CT or multi-modal data, presented comparably excellent performance in COVID-19 discrimination. CONCLUSIONS: The 3DMTM algorithm presented excellent robustness for COVID-19 discrimination with limited CT data. 3DMTM based on CT data performed comparably in COVID-19 discrimination with that trained with multi-modal information. Clinical information could improve the performance of KNN, LR, SVM, and 3DCM in COVID-19 discrimination, especially in the scenario with limited data for training.

Detection and staging of chronic obstructive pulmonary disease using a computed tomography-based weakly supervised deep learning approach.

OBJECTIVES: Chronic obstructive pulmonary disease (COPD) is underdiagnosed globally. The present study aimed to develop weakly supervised deep learning (DL) models that utilize computed tomography (CT) image data for the automated detection and staging of spirometry-defined COPD. METHODS: A large, highly heterogeneous dataset was established, consisting of 1393 participants retrospectively recruited from outpatient, inpatient, and physical examination center settings of four large public hospitals in China. All participants underwent both inspiratory chest CT scans and pulmonary function tests. CT images, spirometry data, demographic information, and clinical information of each participant were collected. An attention-based multi-instance learning (MIL) model for COPD detection was trained using CT scans from 837 participants. External validation of the COPD detection was performed with 620 low-dose CT (LDCT) scans acquired from the National Lung Screening Trial (NLST) cohort. A multi-channel 3D residual network was further developed to categorize GOLD stages among confirmed COPD patients. RESULTS: The attention-based MIL model used for COPD detection achieved an area under the receiver operating characteristic curve (AUC) of 0.934 (95% CI: 0.903, 0.961) on the internal test set and 0.866 (95% CI: 0.805, 0.928) on the LDCT subset acquired from the NLST. The multi-channel 3D residual network was able to correctly grade 76.4% of COPD patients in the test set (423/553) using the GOLD scale. CONCLUSIONS: The proposed chest CT-DL approach can automatically identify spirometry-defined COPD and categorize patients according to the GOLD scale. As such, this approach may be an effective case-finding tool for COPD diagnosis and staging. KEY POINTS: • Chronic obstructive pulmonary disease is underdiagnosed globally, particularly in developing countries. • The proposed chest computed tomography (CT)-based deep learning (DL) approaches could accurately identify spirometry-defined COPD and categorize patients according to the GOLD scale. • The chest CT-DL approach may be an alternative case-finding tool for COPD identification and evaluation.

A deep-learning model for identifying fresh vertebral compression fractures on digital radiography.

OBJECTIVES: To develop a deep-learning (DL) model for identifying fresh VCFs from digital radiography (DR), with magnetic resonance imaging (MRI) as the reference standard. METHODS: Patients with lumbar VCFs were retrospectively enrolled from January 2011 to May 2020. All patients underwent DR and MRI scanning. VCFs were categorized as fresh or old according to MRI results, and the VCF grade and type were assessed. The raw DR data were sent to InferScholar Center for annotation. A DL-based prediction model was built, and its diagnostic performance was evaluated. The DeLong test was applied to assess differences in ROC curves between different models. RESULTS: A total of 1877 VCFs in 1099 patients were included in our study and randomly divided into development (n = 824 patients) and test (n = 275 patients) datasets. The ensemble model identified fresh and old VCFs, reaching an AUC of 0.80 (95% confidence interval [CI], 0.77-0.83), an accuracy of 74% (95% CI, 72-77%), a sensitivity of 80% (95% CI, 77-83%), and a specificity of 68% (95% CI, 63-72%). Lateral (AUC, 0.83) views exhibited better performance than anteroposterior views (AUC, 0.77), and the best performance among respective subgroupings was obtained for grade 3 (AUC, 0.89) and crush-type (AUC, 0.87) subgroups. CONCLUSION: The proposed DL model achieved adequate performance in identifying fresh VCFs from DR. KEY POINTS: • The ensemble deep-learning model identified fresh VCFs from DR, reaching an AUC of 0.80, an accuracy of 74%, a sensitivity of 80%, and a specificity of 68% with the reference standard of MRI. • The lateral views (AUC, 0.83) exhibited better performance than anteroposterior views (AUC, 0.77). • The grade 3 (AUC, 0.89) and crush-type (AUC, 0.87) subgroups showed the best performance among their respective subgroupings.

Diagnosing autism spectrum disorder in children using conventional MRI and apparent diffusion coefficient based deep learning algorithms.

OBJECTIVE: To develop and validate deep learning (DL) methods for diagnosing autism spectrum disorder (ASD) based on conventional MRI (cMRI) and apparent diffusion coefficient (ADC) images. METHODS: A total of 151 ASD children and 151 age-matched typically developing (TD) controls were included in this study. The data from these subjects were assigned to training and validation datasets. An additional 20 ASD children and 25 TD controls were acquired, whose data were utilized in an independent test set. All subjects underwent cMRI and diffusion-weighted imaging examination of the brain. We developed a series of DL models to separate ASD from TD based on the cMRI and ADC data. The seven models used include five single-sequence models (SSMs), one dominant-sequence model (DSM), and one all-sequence model (ASM). To enhance the feature detection of the models, we embed an attention mechanism module. RESULTS: The highest AUC (0.824 ~ 0.850) was achieved when applying the SSM based on either FLAIR or ADC to the validation and independent test sets. A DSM using the combination of FLAIR and ADC showed an improved AUC in the validation (0.873) and independent test sets (0.876). The ASM also showed better diagnostic value in the validation (AUC = 0.838) and independent test sets (AUC = 0.836) compared to the SSMs. Among the models with attention mechanism, the DSM achieved the highest diagnostic performance with an AUC, accuracy, sensitivity, and specificity of 0.898, 84.4%, 85.0%, and 84.0% respectively. CONCLUSIONS: This study established the potential of DL models to distinguish ASD cases from TD controls based on cMRI and ADC images. KEY POINTS: • Deep learning models based on conventional MRI and ADC can be used to diagnose ASD. • The model (DSM) based on the FLAIR and ADC sequence achieved the best diagnostic performance with an AUC of 0.836 in the independent test sets. • The attention mechanism further improved the diagnostic performance of the models.