A deep-learning model for intracranial aneurysm detection on CT angiography images in China: a stepwise, multicentre, early-stage clinical validation study.

BACKGROUND: Artificial intelligence (AI) models in real-world implementation are scarce. Our study aimed to develop a CT angiography (CTA)-based AI model for intracranial aneurysm detection, assess how it helps clinicians improve diagnostic performance, and validate its application in real-world clinical implementation. METHODS: We developed a deep-learning model using 16 546 head and neck CTA examination images from 14 517 patients at eight Chinese hospitals. Using an adapted, stepwise implementation and evaluation, 120 certified clinicians from 15 geographically different hospitals were recruited. Initially, the AI model was externally validated with images of 900 digital subtraction angiography-verified CTA cases (examinations) and compared with the performance of 24 clinicians who each viewed 300 of these cases (stage 1). Next, as a further external validation a multi-reader multi-case study enrolled 48 clinicians to individually review 298 digital subtraction angiography-verified CTA cases (stage 2). The clinicians reviewed each CTA examination twice (ie, with and without the AI model), separated by a 4-week washout period. Then, a randomised open-label comparison study enrolled 48 clinicians to assess the acceptance and performance of this AI model (stage 3). Finally, the model was prospectively deployed and validated in 1562 real-world clinical CTA cases. FINDINGS: The AI model in the internal dataset achieved a patient-level diagnostic sensitivity of 0·957 (95% CI 0·939-0·971) and a higher patient-level diagnostic sensitivity than clinicians (0·943 [0·921-0·961] vs 0·658 [0·644-0·672]; p<0·0001) in the external dataset. In the multi-reader multi-case study, the AI-assisted strategy improved clinicians' diagnostic performance both on a per-patient basis (the area under the receiver operating characteristic curves [AUCs]; 0·795 [0·761-0·830] without AI vs 0·878 [0·850-0·906] with AI; p<0·0001) and a per-aneurysm basis (the area under the weighted alternative free-response receiver operating characteristic curves; 0·765 [0·732-0·799] vs 0·865 [0·839-0·891]; p<0·0001). Reading time decreased with the aid of the AI model (87·5 s vs 82·7 s, p<0·0001). In the randomised open-label comparison study, clinicians in the AI-assisted group had a high acceptance of the AI model (92·6% adoption rate), and a higher AUC when compared with the control group (0·858 [95% CI 0·850-0·866] vs 0·789 [0·780-0·799]; p<0·0001). In the prospective study, the AI model had a 0·51% (8/1570) error rate due to poor-quality CTA images and recognition failure. The model had a high negative predictive value of 0·998 (0·994-1·000) and significantly improved the diagnostic performance of clinicians; AUC improved from 0·787 (95% CI 0·766-0·808) to 0·909 (0·894-0·923; p<0·0001) and patient-level sensitivity improved from 0·590 (0·511-0·666) to 0·825 (0·759-0·880; p<0·0001). INTERPRETATION: This AI model demonstrated strong clinical potential for intracranial aneurysm detection with improved clinician diagnostic performance, high acceptance, and practical implementation in real-world clinical cases. FUNDING: National Natural Science Foundation of China. TRANSLATION: For the Chinese translation of the abstract see Supplementary Materials section.

Low breast density and peritumoral edema on MR predict worse overall survival of breast cancer patients after neoadjuvant chemotherapy.

OBJECTIVES: To investigate the relationship of pre-treatment MR image features (including breast density) and clinical-pathologic characteristics with overall survival (OS) in breast cancer patients receiving neoadjuvant chemotherapy (NAC). METHODS: This retrospective study obtained an approval of the institutional review board and the written informed consents of patients were waived. From October 2013 to April 2019, 130 patients (mean age, 47.6 ± 9.4 years) were included. The univariable and multivariable Cox proportional hazards regression models were applied to analyze factors associated with OS, including MR image parameters and clinical-pathologic characteristics. RESULTS: Among the 130 included patients, 11 (8.5%) patients (mean age, 48.4 ± 11.8 years) died of breast cancer recurrence or distant metastasis. The median follow-up length was 70 months (interquartile range of 60-85 months). According to the Cox regression analysis, older age (hazard ratio [HR] = 1.769, 95% confidence interval [CI]): 1.330, 2.535), higher T stage (HR = 2.490, 95%CI:2.047, 3.029), higher N stage (HR = 1.869, 95%CI:1.507, 2.317), low breast density (HR = 1.693, 95%CI:1.391, 2.060), peritumoral edema (HR = 1.408, 95%CI:1.078, 1.840), axillary lymph nodes invasion (HR = 3.118, 95%CI:2.505, 3.881) on MR were associated with worse OS (all p < 0.05). CONCLUSIONS: Pre-treatment MR features and clinical-pathologic parameters are valuable for predicting long-time OS of breast cancer patients after NAC followed by surgery. Low breast density, peritumoral edema and axillary lymph nodes invasion on pre-treatment MR images were associated with worse prognosis.

Identification of patients with nonischemic dilated cardiomyopathy at risk of malignant ventricular arrhythmias: insights from cardiac magnetic resonance feature tracking.

BACKGROUND: Patients with nonischemic dilated cardiomyopathy (NIDCM) are prone to arrhythmias, and the cause of mortality in these patients is either end-organ dysfunction due to pump failure or malignant arrhythmia-related death. However, the identification of patients with NIDCM at risk of malignant ventricular arrhythmias (VAs) is challenging in clinical practice. The aim of this study was to evaluate whether cardiovascular magnetic resonance feature tracking (CMR-FT) could help in the identification of patients with NIDCM at risk of malignant VAs. METHODS: A total of 263 NIDCM patients who underwent CMR, 24-hour Holter electrocardiography (ECG) and inpatient ECG were retrospectively evaluated. The patients with NIDCM were allocated to two subgroups: NIDCM with VAs and NIDCM without VAs. From CMR-FT, the global peak radial strain (GPRS), global longitudinal strain (GPLS), and global peak circumferential strain (GPCS) were calculated from the left ventricle (LV) model. We investigated the possible predictors of NIDCM combined with VAs by univariate and multivariate logistic regression analyses. RESULTS: The percent LGE (15.51 ± 3.30 vs. 9.62 ± 2.18, P < 0.001) was higher in NIDCM patients with VAs than in NIDCM patients without VAs. Furthermore, the NIDCM patients complicated with VAs had significantly lower GPCS than the NIDCM patients without VAs (- 5.38 (- 7.50, - 4.22) vs.-9.22 (- 10.73, - 8.19), P < 0.01). Subgroup analysis based on LGE negativity showed that NIDCM patients complicated with VAs had significantly lower GPRS, GPCS, and GPLS than NIDCM patients without VAs (P < 0.05 for all). Multivariate analysis showed that both GPCS and %LGE were independent predictors of NIDCM combined with VAs. CONCLUSIONS: CMR global strain can be used to identify NIDCM patients complicated with VAs early, specifically when LGE is not present. GPCS < - 13.19% and %LGE > 10.37% are independent predictors of NIDCM combined with VAs.

HmsU-Net: A hybrid multi-scale U-net based on a CNN and transformer for medical image segmentation.

Accurate medical image segmentation is of great significance for subsequent diagnosis and analysis. The acquisition of multi-scale information plays an important role in segmenting regions of interest of different sizes. With the emergence of Transformers, numerous networks adopted hybrid structures incorporating Transformers and CNNs to learn multi-scale information. However, the majority of research has focused on the design and composition of CNN and Transformer structures, neglecting the inconsistencies in feature learning between Transformer and CNN. This oversight has resulted in the hybrid network's performance not being fully realized. In this work, we proposed a novel hybrid multi-scale segmentation network named HmsU-Net, which effectively fused multi-scale features. Specifically, HmsU-Net employed a parallel design incorporating both CNN and Transformer architectures. To address the inconsistency in feature learning between CNN and Transformer within the same stage, we proposed the multi-scale feature fusion module. For feature fusion across different stages, we introduced the cross-attention module. Comprehensive experiments conducted on various datasets demonstrate that our approach surpasses current state-of-the-art methods.

SaB-Net: Self-attention backward network for gastric tumor segmentation in CT images.

Gastric cancer is a significant contributor to cancer-related fatalities globally. The automated segmentation of gastric tumors has the potential to analyze the medical condition of patients and enhance the likelihood of surgical treatment success. However, the development of an automatic solution is challenged by the heterogeneous intensity distribution of gastric tumors in computed tomography (CT) images, the low-intensity contrast between organs, and the high variability in the stomach shapes and gastric tumors in different patients. To address these challenges, we propose a self-attention backward network (SaB-Net) for gastric tumor segmentation (GTS) in CT images by introducing a self-attention backward layer (SaB-Layer) to feed the self-attention information learned at the deep layer back to the shallow layers. The SaB-Layer efficiently extracts tumor information from CT images and integrates the information into the network, thereby enhancing the network's tumor segmentation ability. We employed datasets from two centers, one for model training and testing and the other for external validation. The model achieved dice scores of 0.8456 on the test set and 0.8068 on the external verification set. Moreover, we validated the model's transfer learning ability on a publicly available liver cancer dataset, achieving results comparable to state-of-the-art liver cancer segmentation models recently developed. SaB-Net has strong potential for assisting in the clinical diagnosis of and therapy for gastric cancer. Our implementation is available at https://github.com/TyrionJ/SaB-Net.

Predicting disease progression in cirrhotic patients after transjugular intrahepatic portosystemic shunt implantation: A sex-stratified analysis.

BACKGROUND: Transjugular intrahepatic portosystemic shunt (TIPS) has been extensively used to treat portal hypertension-associated complications, including cirrhosis. The prediction of post-TIPS prognosis is important for cirrhotic patients, as more aggressive liver transplantation is needed when the post-TIPS prognosis is poor. AIM: To construct a nutrition-based model that could predict the disease progression of cirrhotic patients after TIPS implantation in a sex-dependent manner. METHODS: This study retrospectively recruited cirrhotic patients undergoing TIPS implantation for analysis. Muscle quality was assessed by measuring the skeletal muscle index (SMI) by computed tomography. Multivariate Cox proportional hazard models were utilized to determine the association between SMI and disease progression in cirrhotic patients after TIPS implantation. RESULTS: This study eventually included 186 cirrhotic patients receiving TIPS who were followed up for 30.5 ± 18.8 mo. For male patients, the 30-mo survival rate was significantly lower and the probability of progressive events was higher (3.257-fold) in the low-level SMI group than in the high-level SMI group. According to the multivariate Cox analysis of male patients, SMI < 32.8 was an independent risk factor for long-term adverse outcomes after TIPS implantation. A model was constructed, which involved creatinine, plasma ammonia, SMI, and acute-on-chronic liver failure and hepatic encephalopathy occurring within half a year after surgery. This model had an area under the receiver operating characteristic curve of 0.852, sensitivity of 0.926, and specificity of 0.652. According to the results of the DeLong test, this model outperformed other models (Child-Turcotte-Pugh, Model for End-Stage Liver Disease, and Freiburg index of post-TIPS survival) (P < 0.05). CONCLUSION: SMI is strongly associated with poor long-term outcomes in male patients with cirrhosis who underwent TIPS implantation.

A radiomics based approach using adrenal gland and periadrenal fat CT images to allocate COVID-19 health care resources fairly.

BACKGROUND: The value of radiomics features from the adrenal gland and periadrenal fat CT images for predicting disease progression in patients with COVID-19 has not been studied extensively. We assess the value of radiomics features from the adrenal gland and periadrenal fat CT images in predicting COVID-19 disease exacerbation. METHODS: A total of 1,245 patients (685 moderate and 560 severe patients) were enrolled in a retrospective study. We proposed a 3D V-net to segment adrenal glands in onset CT images automatically, and periadrenal fat was obtained using inflation operation around the adrenal gland. Next, we built a clinical model (CM), three radiomics models (adrenal gland model [AM], periadrenal fat model [PM], and fusion of adrenal gland and periadrenal fat model [FM]), and radiomics nomogram (RN) after radiomics features extracted. RESULTS: The auto-segmentation framework yielded a dice value 0.79 in the training set. CM, AM, PM, FM, and RN obtained AUCs of 0.717, 0.716, 0.736, 0.760, and 0.833 in the validation set. FM and RN had better predictive efficacy than CM (P < 0.0001) in the training set. RN showed that there was no significant difference in the validation set (mean absolute error [MAE] = 0.04) and test set (MAE = 0.075) between predictive and actual results. Decision curve analysis showed that if the threshold probability was between 0.4 and 0.8 in the validation set or between 0.3 and 0.7 in the test set, it could gain more net benefits using RN than FM and CM. CONCLUSIONS: Radiomics features extracted from the adrenal gland and periadrenal fat CT images are related to disease exacerbation in patients with COVID-19.

Case Report: Multidisciplinary management of primary inferior vena cava leiomyosarcoma: a comprehensive case study.

Introduction: Primary Inferior vena cava (IVC) leiomyosarcoma, a rare malignant tumor, presents unique challenges in diagnosis and treatment due to its rarity and the lack of consensus on surgical and adjuvant therapy approaches. Case Report: A 39-year-old female patient presented with lower limb swelling and mild fatigue. Contrast-enhanced CT identified a tumor mass within the dilated IVC. Abdominal MRI revealed primary IVC leiomyosarcoma extending into the right hepatic vein. A multidisciplinary consultation established a diagnosis and devised a treatment plan, opting for Ex-vivo Liver Resection and Auto-transplantation (ELRA), tumor resection and IVC reconstruction. Pathological examination confirmed primary IVC leiomyosarcoma. Postoperatively, the patient underwent a comprehensive treatment strategy that included radiochemotherapy, immunotherapy, targeted therapy, and PRaG therapy (PD-1 inhibitor, Radiotherapy, and Granulocyte-macrophage colony-stimulating factor). Despite the tumor's recurrence and metastasis, the disease progression was partially controlled. Conclusion: This case report emphasizes the complexities of diagnosing and treating IVC leiomyosarcoma and highlights the potential benefits of employing ELRA, IVC reconstruction, and PRaG therapy. Our study may serve as a valuable reference for future investigations addressing the management of this rare disease.

A robust and efficient AI assistant for breast tumor segmentation from DCE-MRI via a spatial-temporal framework.

Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) allows screening, follow up, and diagnosis for breast tumor with high sensitivity. Accurate tumor segmentation from DCE-MRI can provide crucial information of tumor location and shape, which significantly influences the downstream clinical decisions. In this paper, we aim to develop an artificial intelligence (AI) assistant to automatically segment breast tumors by capturing dynamic changes in multi-phase DCE-MRI with a spatial-temporal framework. The main advantages of our AI assistant include (1) robustness, i.e., our model can handle MR data with different phase numbers and imaging intervals, as demonstrated on a large-scale dataset from seven medical centers, and (2) efficiency, i.e., our AI assistant significantly reduces the time required for manual annotation by a factor of 20, while maintaining accuracy comparable to that of physicians. More importantly, as the fundamental step to build an AI-assisted breast cancer diagnosis system, our AI assistant will promote the application of AI in more clinical diagnostic practices regarding breast cancer.

TGMIL: A hybrid multi-instance learning model based on the Transformer and the Graph Attention Network for whole-slide images classification of renal cell carcinoma.

BACKGROUND AND OBJECTIVES: The pathological diagnosis of renal cell carcinoma is crucial for treatment. Currently, the multi-instance learning method is commonly used for whole-slide image classification of renal cell carcinoma, which is mainly based on the assumption of independent identical distribution. But this is inconsistent with the need to consider the correlation between different instances in the diagnosis process. Furthermore, the problem of high resource consumption of pathology images is still urgent to be solved. Therefore, we propose a new multi-instance learning method to solve this problem. METHODS: In this study, we proposed a hybrid multi-instance learning model based on the Transformer and the Graph Attention Network, called TGMIL, to achieve whole-slide image of renal cell carcinoma classification without pixel-level annotation or region of interest extraction. Our approach is divided into three steps. First, we designed a feature pyramid with the multiple low magnifications of whole-slide image named MMFP. It makes the model incorporates richer information, and reduces memory consumption as well as training time compared to the highest magnification. Second, TGMIL amalgamates the Transformer and the Graph Attention's capabilities, adeptly addressing the loss of instance contextual and spatial. Within the Graph Attention network stream, an easy and efficient approach employing max pooling and mean pooling yields the graph adjacency matrix, devoid of extra memory consumption. Finally, the outputs of two streams of TGMIL are aggregated to achieve the classification of renal cell carcinoma. RESULTS: On the TCGA-RCC validation set, a public dataset for renal cell carcinoma, the area under a receiver operating characteristic (ROC) curve (AUC) and accuracy of TGMIL were 0.98±0.0015,0.9191±0.0062, respectively. It showcased remarkable proficiency on the private validation set of renal cell carcinoma pathology images, attaining AUC of 0.9386±0.0162 and ACC of 0.9197±0.0124. Furthermore, on the public breast cancer whole-slide image test dataset, CAMELYON 16, our model showed good classification performance with an accuracy of 0.8792. CONCLUSIONS: TGMIL models the diagnostic process of pathologists and shows good classification performance on multiple datasets. Concurrently, the MMFP module efficiently diminishes resource requirements, offering a novel angle for exploring computational pathology images.

Deep wavelet scattering orthogonal fusion network for glioma IDH mutation status prediction.

Accurately predicting the isocitrate dehydrogenase (IDH) mutation status of gliomas is greatly significant for formulating appropriate treatment plans and evaluating the prognoses of gliomas. Although existing studies can accurately predict the IDH mutation status of gliomas based on multimodal magnetic resonance (MR) images and machine learning methods, most of these methods cannot fully explore multimodal information and effectively predict IDH status for datasets acquired from multiple centers. To address this issue, a novel wavelet scattering (WS)-based orthogonal fusion network (WSOFNet) was proposed in this work to predict the IDH mutation status of gliomas from multiple centers. First, transformation-invariant features were extracted from multimodal MR images with a WS network, and then the multimodal WS features were used instead of the original images as the inputs of WSOFNet and were fully fused through an adaptive multimodal feature fusion module (AMF2M) and an orthogonal projection module (OPM). Finally, the fused features were input into a fully connected classifier to predict IDH mutation status. In addition, to achieve improved prediction accuracy, four auxiliary losses were also used in the feature extraction modules. The comparison results showed that the prediction area under the curve (AUC) of WSOFNet on a single-center dataset was 0.9966 and that on a multicenter dataset was approximately 0.9655, which was at least 3.9% higher than that of state-of-the-art methods. Moreover, the ablation experimental results also proved that the adaptive multimodal feature fusion strategy based on orthogonal projection could effectively improve the prediction performance of the model, especially for an external validation dataset.

Abdominal pregnancy secondary to uterine horn pregnancy: a case report.

BACKGROUND: Pregnancy begins with a fertilized ovum that normally attaches to the uterine endometrium. However, an ectopic pregnancy can occur when a fertilized egg implants and grows outside the uterine cavity. Tubal ectopic pregnancy is the most common type (over 95%), with ovarian, abdominal, cervical, broad ligament, and uterine cornual pregnancy being less common. As more cases of ectopic pregnancy are diagnosed and treated in the early stages, the survival rate and fertility retention significantly improve. However, complications of abdominal pregnancy can sometimes be life-threatening and have severe consequences. CASE PRESENTATION: We present a case of intraperitoneal ectopic pregnancy with fetal survival. Ultrasound and magnetic resonance imaging showed a right cornual pregnancy with a secondary abdominal pregnancy. In September 2021, we performed an emergency laparotomy, along with additional procedures such as transurethral ureteroscopy, double J-stent placement, abdominal fetal removal, placentectomy, repair of the right uterine horn, and pelvic adhesiolysis, in the 29th week of pregnancy. During laparotomy, we diagnosed abdominal pregnancy secondary to a rudimentary uterine horn. The mother and her baby were discharged eight days and 41 days, respectively, after surgery. CONCLUSIONS: Abdominal pregnancy is a rare condition. The variable nature of ectopic pregnancy can cause delays in timely diagnosis, resulting in increased morbidity and mortality, especially in areas with inadequate medical and social services. A high index of suspicion, coupled with appropriate imaging studies, can help facilitate its diagnosis in any suspected case.

msQSM: Morphology-based self-supervised deep learning for quantitative susceptibility mapping.

Quantitative susceptibility mapping (QSM) has been applied to the measurement of iron deposition and the auxiliary diagnosis of neurodegenerative disease. There still exists a dipole inversion problem in QSM reconstruction. Recently, deep learning approaches have been proposed to resolve this problem. However, most of these approaches are supervised methods that need pairs of the input phase and ground-truth. It remains a challenge to train a model for all resolutions without using the ground-truth and only using one resolution data. To address this, we proposed a self-supervised QSM deep learning method based on morphology. It consists of a morphological QSM builder to decouple the dependency of the QSM on acquisition resolution, and a morphological loss to reduce artifacts effectively and save training time efficiently. The proposed method can reconstruct arbitrary resolution QSM on both human data and animal data, regardless of whether the resolution is higher or lower than that of the training set. Our method outperforms the previous best unsupervised method with a 3.6% higher peak signal-to-noise ratio, 16.2% lower normalized root mean square error, and 22.1% lower high-frequency error norm. The morphological loss reduces training time by 22.1% with respect to the cycle gradient loss used in the previous unsupervised methods. Experimental results show that the proposed method accurately measures QSM with arbitrary resolutions, and achieves state-of-the-art results among unsupervised deep learning methods. Research on applications in neurodegenerative diseases found that our method is robust enough to measure significant increase in striatal magnetic susceptibility in patients during Alzheimer's disease progression, as well as significant increase in substantia nigra susceptibility in Parkinson's disease patients, and can be used as an auxiliary differential diagnosis tool for Alzheimer's disease and Parkinson's disease.

Novel compound heterozygous mutations in the AFG3L2 gene in a Chinese child with microcephaly, early-onset seizures, and cerebral atrophy.

Background: The most common disease caused by biallelic AFG3L2 mutations is spastic ataxia type 5 (SPAX5). Identification of complex phenotypes resulting from biallelic AFG3L2 mutations has been increasing in recent years. Methods: A retrospective analysis was performed on a child with microcephaly and recurrent seizures. The child underwent physical and neurological examinations, laboratory tests, electroencephalography (EEG), and brain magnetic resonance imaging (MRI). Trio-whole-exome sequencing (trio-WES) was performed to identify possible causative mutations. Results: We described a child who exhibited early-onset and intractable epilepsy, developmental regression, microcephaly, and premature death. Neuroimaging revealed global cerebral atrophy (GCA) involving the cerebrum, cerebellum, corpus callosum, brainstem, cerebellar vermis, and basal ganglia. On trio-WES, two novel compound heterozygous mutations, c.1834G > T (p.E612*) and c.2176-6T > A in the AFG3L2 gene, were identified in this patient. Conclusions: Our findings have expanded the mutation spectrum of the AFG3L2 gene and identified a severe neurodegenerative phenotype of global cerebral atrophy caused by biallelic AFG3L2 mutations.

Evaluation of condylar osseous changes using a wireless detector with proton density-weighted imaging sequences.

Background: Cone-beam computed tomography (CBCT) is the gold standard for evaluating condylar osseous changes. However, the radiation risk and low soft-tissue resolution of CBCT make it unsuitable for evaluating soft tissue such as the articular disc and lateral pterygoid muscle. This study aimed to qualitatively and quantitatively evaluate the feasibility and advantages of using wireless detectors (WD) with proton density-weighted imaging (PDWI) sequences to image condyle changes in patients with temporomandibular disorders (TMD). Methods: This study involved 20 patients (male =8, female =12; mean age 31.65 years, SD 12.68 years) with TMD. All participants underwent a closed oblique sagittal PDWI scan with head/neck coupling coiling (HNCC) and wireless detector-HNCC (WD-HNCC) on a 3.0 T magnetic resonance imaging (MRI) scanner. Subsequently, the changes in the condyle bones in the scanned images for the 2 image types were scored subjectively and compared qualitatively. The contrast-to-noise ratio (CNR) of the 2 types of scanned images was compared quantitatively. The comparison of CNR differences between the 2 types of images was performed using the paired t-test. The kappa statistic was used to test the consistency of quantitative analyses of MRI images between observers. The subjective scores of condylar osseous changes in the 2 types of images were compared by paired rank-sum test. A P value <0.05 was considered statistically significant. Results: A total of 40 condyles from 20 patients were scanned. Among them, 8 condyles showed no bone changes, and the other 32 condyles demonstrated condylar osseous changes of varying degrees and nature. These 32 condyles were used in the subsequent analysis. As compared to images acquired by HNCC in the PDWI sequence, the WD-HNCC images more clearly showed mandibular osteophyte, bone cortical erosion, subcortical cystic focus, and bone cortical hyperplasia and thickening. In addition, the WD-HNCC was demonstrated to improve image CNR by 158.9% compared to HNCC (28.17±16.01 vs. 10.88±6.53; t=8.63; P=0.001). Conclusions: WD-HNCC in PDWI sequences is suitable for imaging the condylar bone changes of patients with TMD and significantly improves the image quality.

Changes in Volume of Subregions Within Basal Ganglia in Obsessive-Compulsive Disorder: A Study With Atlas-Based and VBM Methods.

Background: The role of basal ganglia in the pathogenesis of obsessive-compulsive disorder (OCD) remains unclear. The studies on volume changes of basal ganglia in OCD commonly use the VBM method; however, the Atlas-based method used in such research has not been reported. Atlas-based method has a lower false positive rate compared with VBM method, thus having advantages partly. Objectives: The current study aimed to detect the volume changes of subregions within basal ganglia in OCD using Atlas-based method to further delineate the precise neural circuitry of OCD. What is more, we explored the influence of software used in Atlas-based method on the volumetric analysis of basal ganglia and compared the results of Atlas-based method and regularly used VBM method. Methods: We analyzed the brain structure images of 37 patients with OCD and 41 healthy controls (HCs) using the VBM method, Atlas-based method based on SPM software, or Freesurfer software to find the areas with significant volumetric variation between the two groups, and calculated the effects size of these areas. Results: VBM analysis revealed a significantly increased volume of bilateral lenticular nucleus in patients compared to HCs. In contrast, Atlas-based method based on Freesurfer revealed significantly increased volume of left globus pallidus in patients, and the largest effect size of volumetric variation was revealed by Freesurfer analysis. Conclusions: This study showed that the volume of bilateral lenticular nucleus significantly increased in patients compared to HCs, especially left globus pallidus, which was in accordance with the previous findings. In addition, Freesurfer is better than SPM and a good choice for Atlas-based volumetric analysis of basal ganglia.

StoHisNet: A hybrid multi-classification model with CNN and Transformer for gastric pathology images.

BACKGROUND AND OBJECTIVES: Gastric cancer has high morbidity and mortality compared to other cancers. Accurate histopathological diagnosis has great significance for the treatment of gastric cancer. With the development of artificial intelligence, many researchers have applied deep learning for the classification of gastric cancer pathological images. However, most studies have used binary classification on pathological images of gastric cancer, which is insufficient with respect to the clinical requirements. Therefore, we proposed a multi-classification method based on deep learning with more practical clinical value. METHODS: In this study, we developed a novel multi-scale model called StoHisNet based on Transformer and the convolutional neural network (CNN) for the multi-classification task. StoHisNet adopts Transformer to learn global features to alleviate the inherent limitations of the convolution operation. The proposed StoHisNet can classify the publicly available pathological images of a gastric dataset into four categories -normal tissue, tubular adenocarcinoma, mucinous adenocarcinoma, and papillary adenocarcinoma. RESULTS: The accuracy, F1-score, recall, and precision of the proposed model in the public gastric pathological image dataset were 94.69%, 94.96%, 94.95%, and 94.97%, respectively. We conducted additional experiments using two other public datasets to verify the generalization ability of the model. On the BreakHis dataset, our model performed better compared with other classification models, and the accuracy was 91.64%. Similarly, on the four-classification task on the Endometrium dataset, our model showed better classification ability than others with accuracy of 81.74%. These experiments showed that the proposed model has excellent ability of classification and generalization. CONCLUSION: The StoHisNet model had high performance in the multi-classification on gastric histopathological images and showed strong generalization ability on other pathological datasets. This model may be a potential tool to assist pathologists in the analysis of gastric histopathological images.

Learn Less, Infer More: Learning in the Fourier Domain for Quantitative Susceptibility Mapping.

Quantitative susceptibility mapping (QSM) aims to evaluate the distribution of magnetic susceptibility from magnetic resonance phase measurements by solving the ill-conditioned dipole inversion problem. Removing the artifacts and preserving the anisotropy of tissue susceptibility simultaneously is still a challenge in QSM. To deal with this issue, a novel k-QSM network is proposed to resolve dipole inversion issues in QSM reconstruction. The k-QSM network converts the results obtained by truncated k-space division (TKD) into the Fourier domain as inputs. After passing through several convolutional and residual blocks, the ill-posed signals of TKD are corrected by making the network output close to the calculation of susceptibility through multiple orientation sampling (COSMOS)-labeled QSM. To evaluate the superiority of k-QSM, comparisons with several state-of-the-art methods are performed in terms of QSM artifacts removing, anisotropy preserving, generalization ability, and clinical applications. Compared to existing methods, the k-QSM achieves a 22.31% lower normalized root mean square error, 10.30% higher peak signal-to-noise ratio (PSNR), 33.10% lower high-frequency error norm, and 1.06% higher structural similarity. In addition, the orientation-dependent susceptibility variation obtained by k-QSM is significant, verifying that k-QSM has the ability to preserve susceptibility anisotropy. When the trained models are tested on the dataset from different centers, our k-QSM shows a strong generalization ability with the highest PSNR. Moreover, by comparing the susceptibility maps between healthy controls and drug addicts with different methods, we found the proposed k-QSM is more sensitive to the susceptibility abnormality in the patients. The proposed k-QSM method learns less-only to fix the ill-posed signals of TKD, but infers more-both COSMOS-like and anisotropy-preserving QSM results. Its generalization ability and great sensitivity to susceptibility changes can make it a potential method for distinguishing some diseases.