An Automated Deep Learning-Based Framework for Uptake Segmentation and Classification on PSMA PET/CT Imaging of Patients with Prostate Cancer

Uptake segmentation and classification on PSMA PET/CT are important for automating whole-body tumor burden determinations. We developed and evaluated an automated deep learning (DL)-based framework that segments and classifies uptake on PSMA PET/CT. We identified 193 [18F] DCFPyL PET/CT scans of patients with biochemically recurrent prostate cancer from two institutions, including 137 [18F] DCFPyL PET/CT scans for training and internally testing, and 56 scans from another institution for external testing. Two radiologists segmented and labelled foci as suspicious or non-suspicious for malignancy. A DL-based segmentation was developed with two independent CNNs. An anatomical prior guidance was applied to make the DL framework focus on PSMA-avid lesions. Segmentation performance was evaluated by Dice, IoU, precision, and recall. Classification model was constructed with multi-modal decision fusion framework evaluated by accuracy, AUC, F1 score, precision, and recall. Automatic segmentation of suspicious lesions was improved under prior guidance, with mean Dice, IoU, precision, and recall of 0.700, 0.566, 0.809, and 0.660 on the internal test set and 0.680, 0.548, 0.749, and 0.740 on the external test set. Our multi-modal decision fusion framework outperformed single-modal and multi-modal CNNs with accuracy, AUC, F1 score, precision, and recall of 0.764, 0.863, 0.844, 0.841, and 0.847 in distinguishing suspicious and non-suspicious foci on the internal test set and 0.796, 0.851, 0.865, 0.814, and 0.923 on the external test set. DL-based lesion segmentation on PSMA PET is facilitated through our anatomical prior guidance strategy. Our classification framework differentiates suspicious foci from those not suspicious for cancer with good accuracy.

Classification of recurrent major depressive disorder using a residual denoising autoencoder framework: Insights from large-scale multisite fMRI data

BACKGROUND AND OBJECTIVE

Recurrent major depressive disorder (rMDD) has a high recurrence rate, and symptoms often worsen with each episode. Classifying rMDD using functional magnetic resonance imaging (fMRI) can enhance understanding of brain activity and aid diagnosis and treatment of this disorder.

METHODS

We developed a Residual Denoising Autoencoder (Res-DAE) framework for the classification of rMDD. The functional connectivity (FC) was extracted from fMRI data as features. The framework addresses site heterogeneity by employing the Combat method to harmonize feature distribution differences. A feature selection method based on Fisher scores was used to reduce redundant information in the features. A data augmentation strategy using a Synthetic Minority Over-sampling Technique algorithm based on Extended Frobenius Norm measure was incorporated to increase the sample size. Furthermore, a residual module was integrated into the autoencoder network to preserve important features and improve the classification accuracy.

RESULTS

We tested our framework on a large-scale, multisite fMRI dataset, which includes 189 rMDD patients and 427 healthy controls. The Res-DAE achieved an average accuracy of 75.1 % (sensitivity = 69 %, specificity = 77.8 %) in cross-validation, thereby outperforming comparison methods. In a larger dataset that also includes first-episode depression (comprising 832 MDD patients and 779 healthy controls), the accuracy reached 70 %.

CONCLUSIONS

We proposed a deep learning framework that can effectively classify rMDD and 33 identify the altered FC associated with rMDD. Our study may reveal changes in brain function 34 associated with rMDD and provide assistance for the diagnosis and treatment of rMDD.

A novel approach for intelligent diagnosis and grading of diabetic retinopathy

Diabetic retinopathy (DR) is a severe ocular complication of diabetes that can lead to vision damage and even blindness. Currently, traditional deep convolutional neural networks (CNNs) used for DR grading tasks face two primary challenges: (1) insensitivity to minority classes due to imbalanced data distribution, and (2) neglecting the relationship between the left and right eyes by utilizing the fundus image of only one eye for training without differentiating between them. To tackle these challenges, we proposed the DRGCNN (DR Grading CNN) model. To solve the problem caused by imbalanced data distribution, our model adopts a more balanced strategy by allocating an equal number of channels to feature maps representing various DR categories. Furthermore, we introduce a CAM-EfficientNetV2-M encoder dedicated to encoding input retinal fundus images for feature vector generation. The number of parameters of our encoder is 52.88 M, which is less than RegNet_y_16gf (80.57 M) and EfficientNetB7 (63.79 M), but the corresponding kappa value is higher. Additionally, in order to take advantage of the binocular relationship, we input fundus retinal images from both eyes of the patient into the network for features fusion during the training phase. We achieved a kappa value of 86.62% on the EyePACS dataset and 86.16% on the Messidor-2 dataset. Experimental results on these representative datasets for diabetic retinopathy (DR) demonstrate the exceptional performance of our DRGCNN model, establishing it as a highly competitive intelligent classification model in the field of DR. The code is available for use at https://github.com/Fat-Hai/DRGCNN.

Active Learning in Brain Tumor Segmentation with Uncertainty Sampling and Annotation Redundancy Restriction

Deep learning models have demonstrated great potential in medical imaging but are limited by the expensive, large volume of annotations required. To address this, we compared different active learning strategies by training models on subsets of the most informative images using real-world clinical datasets for brain tumor segmentation and proposing a framework that minimizes the data needed while maintaining performance. Then, 638 multi-institutional brain tumor magnetic resonance imaging scans were used to train three-dimensional U-net models and compare active learning strategies. Uncertainty estimation techniques including Bayesian estimation with dropout, bootstrapping, and margins sampling were compared to random query. Strategies to avoid annotating similar images were also considered. We determined the minimum data necessary to achieve performance equivalent to the model trained on the full dataset (α = 0.05). Bayesian approximation with dropout at training and testing showed results equivalent to that of the full data model (target) with around 30% of the training data needed by random query to achieve target performance (p = 0.018). Annotation redundancy restriction techniques can reduce the training data needed by random query to achieve target performance by 20%. We investigated various active learning strategies to minimize the annotation burden for three-dimensional brain tumor segmentation. Dropout uncertainty estimation achieved target performance with the least annotated data.

DGCBG-Net: A dual-branch network with global cross-modal interaction and boundary guidance for tumor segmentation in PET/CT images

BACKGROUND AND OBJECTIVES

Automatic tumor segmentation plays a crucial role in cancer diagnosis and treatment planning. Computed tomography (CT) and positron emission tomography (PET) are extensively employed for their complementary medical information. However, existing methods ignore bilateral cross-modal interaction of global features during feature extraction, and they underutilize multi-stage tumor boundary features.

METHODS

To address these limitations, we propose a dual-branch tumor segmentation network based on global cross-modal interaction and boundary guidance in PET/CT images (DGCBG-Net). DGCBG-Net consists of 1) a global cross-modal interaction module that extracts global contextual information from PET/CT images and promotes bilateral cross-modal interaction of global feature; 2) a shared multi-path downsampling module that learns complementary features from PET/CT modalities to mitigate the impact of misleading features and decrease the loss of discriminative features during downsampling; 3) a boundary prior-guided branch that extracts potential boundary features from CT images at multiple stages, assisting the semantic segmentation branch in improving the accuracy of tumor boundary segmentation.

RESULTS

Extensive experiments are conducted on STS and Hecktor 2022 datasets to evaluate the proposed method. The average Dice scores of our DGCB-Net on the two datasets are 80.33% and 79.29%, with average IOU scores of 67.64% and 70.18%. DGCB-Net outperformed the current state-of-the-art methods with a 1.77% higher Dice score and a 2.12% higher IOU score.

CONCLUSIONS

Extensive experimental results demonstrate that DGCBG-Net outperforms existing segmentation methods, and is competitive to state-of-arts.

Classification of MDD using a Transformer classifier with large-scale multisite resting-state fMRI data

Major depressive disorder (MDD) is one of the most common psychiatric disorders worldwide with high recurrence rate. Identifying MDD patients, particularly those with recurrent episodes with resting-state fMRI, may reveal the relationship between MDD and brain function. We proposed a Transformer-Encoder model, which utilized functional connectivity extracted from large-scale multisite rs-fMRI datasets to classify MDD and HC. The model discarded the Transformer's Decoder part, reducing the model's complexity and decreasing the number of parameters to adapt to the limited sample size and it does not require a complex feature selection process and achieves end-to-end classification. Additionally, our model is suitable for classifying data combined from multiple brain atlases and has an optional unsupervised pre-training module to acquire optimal initial parameters and speed up the training process. The model's performance was tested on a large-scale multisite dataset and identified brain regions affected by MDD using the Grad-CAM method. After conducting five-fold cross-validation, our model achieved an average classification accuracy of 68.61% on a dataset consisting of 1611 samples. For the selected recurrent MDD dataset, the model reached an average classification accuracy of 78.11%. Abnormalities were detected in the frontal gyri and cerebral cortex of MDD patients in both datasets. Furthermore, the identified brain regions in the recurrent MDD dataset generally exhibited a higher contribution to the model's performance.

TGRA-P: Task-driven model predicts 90-day mortality from ICU clinical notes on mechanical ventilation

BACKGROUND

With the outbreak and spread of COVID-19 worldwide, limited ventilators fail to meet the surging demand for mechanical ventilation in the ICU. Clinical models based on structured data that have been proposed to rationalize ventilator allocation often suffer from poor ductility due to fixed fields and laborious normalization processes. The advent of pre-trained models and downstream fine-tuning methods allows for learning large amounts of unstructured clinical text for different tasks. But the hardware requirements of large-scale pre-trained models and purposeless networks downstream have led to a lack of promotion in the clinical domain.

OBJECTIVE

In this study, an innovative architecture of a task-driven predictive model is proposed and a Task-driven Gated Recurrent Attention Pool model (TGRA-P) is developed based on the architecture. TGRA-P predicts early mortality risk from patients' clinical notes on mechanical ventilation in the ICU, which is used to assist clinicians in diagnosis and decision-making.

METHODS

Specifically, a Task-Specific Embedding Module is proposed to fine-tune the embedding with task labels and save it as static files for downstream calls. It serves the task better and prevents GPU overload. The Gated Recurrent Attention Unit (GRA) is proposed to further enhance the dependency of the information preceding and following the text sequence with fewer parameters. In addition, we propose a Residual Max Pool (RMP) to avoid ignoring words in common text classification tasks by incorporating all word-level features of the notes for prediction. Finally, we use a fully connected decoding network as a classifier to predict the mortality risk.

RESULT

The proposed model shows very promising results with an AUROC of 0.8245±0.0096, an AUPRC of 0.7532±0.0115, an accuracy of 0.7422±0.0028 and F1-score of 0.6612±0.0059 for 90-day mortality prediction using clinical notes of ICU mechanically ventilated patients on the MIMIC-III dataset, all of which are better than previous studies. Moreover, the superiority of the proposed model in comparison with other baseline models is also statistically validated through the calculated Cohen's d effect sizes.

CONCLUSION

The experimental results show that TGRA-P based on the innovative task-driven prognostic architecture obtains state-of-the-art performance. In future work, we will build upon the provided code and investigate its applicability to different datasets. The model balances performance and efficiency, not only reducing the cost of early mortality risk prediction but also assisting physicians in making timely clinical interventions and decisions. By incorporating textual records that are challenging for clinicians to utilize, the model serves as a valuable complement to physicians' judgment, enhancing their decision-making process.

Classification of recurrent major depressive disorder using a new time series feature extraction method through multisite rs-fMRI data

BACKGROUND

Major depressive disorder (MDD) has a high rate of recurrence. Identifying patients with recurrent MDD is advantageous in adopting prevention strategies to reduce the disabling effects of depression.

METHOD

We propose a novel feature extraction method that includes dynamic temporal information, and inputs the extracted features into a graph convolutional network (GCN) to achieve classification of recurrent MDD. We extract the average time series using an atlas from resting-state functional magnetic resonance imaging (fMRI) data. Pearson correlation was calculated between brain region sequences at each time point, representing the functional connectivity at each time point. The connectivity is used as the adjacency matrix and the brain region sequences as node features for a GCN model to classify recurrent MDD. Gradient-weighted Class Activation Mapping (Grad-CAM) was used to analyze the contribution of different brain regions to the model. Brain regions making greater contribution to classification were considered to be the regions with altered brain function in recurrent MDD.

RESULT

We achieved a classification accuracy of 75.8 % for recurrent MDD on the multi-site dataset, the Rest-meta-MDD. The brain regions closely related to recurrent MDD have been identified.

LIMITATION

The pre-processing stage may affect the final classification performance and harmonizing site differences may improve the classification performance.

CONCLUSION

The experimental results demonstrate that the proposed method can effectively classify recurrent MDD and extract dynamic changes of brain activity patterns in recurrent depression.

PD-1 Inhibitors and Chemotherapy Combined with or without Radiotherapy for Patients with Oligometastatic Esophageal Cancer

PURPOSE/OBJECTIVE(S)

Oligometastatic esophageal cancer (OMEC) is an intermediate state between local advanced and widespread metastatic disease, which is associated with better prognosis compared to poly-metastatic esophageal cancer (EC). The previous studies demonstrated the survival benefit from local radiotherapy for OMEC patients. But the data of PD-1 inhibitors combined with radiotherapy for OMEC is still scarce. The purpose of the present study was to determine the efficacy and safety of PD-1 inhibitors plus radiotherapy in OMEC.

MATERIALS/METHODS

OMEC was defined as "up to five measurable metastatic lesions and up to three organs involved". Patients with OMEC receiving PD-1 inhibitors plus chemotherapy in a single center were retrospectively analyzed in this study. They were dichotomized according to whether or not they had received radiotherapy. The efficacy and safety of immunochemotherapy combined with radiotherapy (RT group) and immunochemotherapy alone (NRT group) were investigated.

RESULTS

A total of 226 patients were included; 108 patients received PD-1 inhibitors plus chemotherapy and radiotherapy, while other 118 patients were treated with immunochemotherapy alone. Baseline characteristics were well balanced between the groups. The overall response rate (ORR) was 58.3% in the RT group and 41.5% in the NRT group (P = 0.012), respectively. The median PFS was 13.5 months (95% CI, 10.0-17.1) for the RT group and 8.8 months (95% CI, 9.2-12.0) for the NRT group (P = 0.000). The addition of radiotherapy was the major prognostic factor for PFS (hazard ratio, 0.56; 95% CI, 0.406-0.761; P = 0.000) by univariate Cox regression analysis. Patients were well-tolerated, and the overall incidence of adverse events was similar between the RT group and NRT group. In addition, the incidence of treatment-related pneumonitis did not differ between the two groups. Grade 3-5 pneumonitis was observed in 3.7% and 5.1% of patients in the RT and NRT groups, respectively.

CONCLUSION

The additional of radiotherapy to PD-1 inhibitors and chemotherapy improved PFS of patients with OMEC and showed acceptable toxicity. Further prospective studies investigating the combination of immunochemotherapy and radiotherapy are warranted.

Safety and Efficacy Analysis of Patients with Extensive-Stage Small Cell Lung Cancer (ES-SCLC) Treated with SHR-1316 Plus Chemotherapy and Sequential Chest Radiotherapy as First-Line Therapy from a Phase II Trial

PURPOSE/OBJECTIVE(S)

CAPSTONE-1, a phase 3 trial, showed that SHR-1316 (PD-L1 antibody) combined with standard first-line chemotherapy could prolong overall survival (OS) in patients (pts) with ES-SCLC. The CREST trial reported consolidative thoracic radiotherapy (TRT) of 30 Gy in 10 fractions provided a 10% 2-year OS benefit and more intensive TRT should be investigated in ES-SCLC. In the era of immunotherapy, the role of TRT also needs further exploration. Therefore, we designed this clinical trial to investigate the efficacy and safety of SHR-1316 plus first-line chemotherapy followed by TRT combined with SHR-1316.

MATERIALS/METHODS

Key inclusion criteria were pts aged 18-75 years, with previously untreated histologically or cytologically confirmed ES-SCLC, and an ECOG performance status of 0-1. Eligible pts would receive 4∼6 cycles of SHR-1316 (20mg/kg, D1, q3w) combined with EP/EC (etoposide, 100mg/m2, D1-5, q3w and cisplatin, 75mg/m², D1-3, q3w or carboplatin, AUC = 5, D1, q3w), followed by SHR-1316 combined with TRT (≥3 Gy*10 f or ≥2 Gy*25 f, involved-field irradiation), and then the maintenance therapy with SHR-1316 until disease progression or intolerable adverse events (AEs). The main endpoints included ORR, PFS and safety.

RESULTS

From October 2020 to January 2023, 33 pts received SHR-1316 and sequential consolidative TRT. Among them, 19 pts received high-dose TRT (>3 Gy*10 f or ≥2 Gy*25 f) and 14 pts received low-dose TRT (≤3 Gy*10 f or<2 Gy*25 f). The median age was 62 (range: 38-73). Most pts were male (28, 84.8%), former smokers (22, 66.7%) with an ECOG performance status 1 (32, 97%). Ten (30.3%) pts were diagnosed with brain metastasis and 10 (30.3%) pts had liver metastasis at baseline. At the data cutoff date, 9 pts remained on treatment, the average number of treatment cycles was 9.2. 33 pts had at least one 1 post-treatment tumor assessment. The confirmed ORR and DCR were 90.9% (30/33) and 100% (33/33) in all pts, were 89.5% (17/19) and 100% (19/19) in high-dose TRT group, and were 92.9% (13/14) and 100% (14/14) in low-dose TRT group. The median PFS was 10.2(CI: 5.8∼14.7) months in all pts, was 7 (CI: 3.8∼10.2) months in high-dose TRT group and 10.4 (CI: 8.4∼12.3) months in low-dose TRT group. AEs occurred in 27 (81.8%) pts and grade 3 or 4 AEs occurred in 20 (60.6%) pts. The most common grade 3 or 4 AEs included neutropenia (15, 45.5%), leukopenia (8, 24.2%), lymphocytopenia (5, 15.2%), pneumonia (3, 9.1%), anemia (3, 9.1%) and thrombocytopenia (2, 6.1%).

CONCLUSION

SHR-1316 plus chemotherapy and sequential TRT as first-line therapy for ES-SCLC showed promising efficacy and acceptable safety. There is no significant difference between high-dose and low-dose TRT groups in terms of safety and efficacy according to current data.

Thoracic Radiotherapy Improves the Outcomes of Extensive Stage Small-Cell Lung Cancer Patients Receiving First-Line Immunotherapy: A Multicenter Retrospective Analysis

PURPOSE/OBJECTIVE(S)

Platinum-etoposide chemotherapy combined with immune checkpoint inhibitors has been recommended as the first line standard treatment for extensive-stage small-cell lung cancer (ES-SCLC). However, the role of thoracic radiotherapy (TRT) was still unknown for these patients. The aim of this study was to evaluate the efficacy and safety of TRT for ES-SCLC patients receiving first-line immunotherapy and chemotherapy MATERIALS/METHODS: ES-SCLC patients who received 4 to 6 cycles of chemotherapy and immunotherapy as first-line therapy from two hospitals were included in analysis between July 2018 and January 2023. All patients were divided into two groups based on whether receiving TRT or not during the first-line setting. The primary endpoints were overall survival (OS) and progression-free survival (PFS), and the secondary endpoint was toxic effects. The Kaplan-Meier method was used to estimate overall survival and progression-free survival. All adverse events were graded by the senior doctors according to the Common Terminology Criteria for Adverse Events version 5.0.

RESULTS

A total of 253 patients from two hospitals were enrolled in analysis. The median age was 62 years and most patients were men (83%), and 36% patients were staged T4 and 52% N3. The most common sites of metastasis were brain (32%), liver (32%) and bone (30%). There were 107 patients (42%) receiving TRT and 146 (58%) without TRT. Baseline characteristics were well balanced between the two groups. The median follow-up time was 16.7 months. Statistically significant benefit was observed for patients receiving TRT compared to patients without TRT (median PFS, 10.4 vs 7.3 months, P< 0.001; median OS, 22.2 vs 13.7 months, P = 0.009). In terms of safety, no significant increase of any grade adverse event (AE) (P = 0.115) and grade 3 or 4 AE (P = 0.631) were observed for patients receiving TRT. The most common grade 3 or 4 AE were neutrophil count decreased, white blood cell count decreased, and nausea in the two groups. In the TRT group, the most common grade 1 or 2 AE related to TRT were esophagitis (40%) and pneumonitis (25%). Grade 3 or 4 esophagitis and pneumonitis were 4% and 8%, respectively. Only one patient developed grade 4 toxic effect of pneumonitis leading to radiotherapy withdrawal. No grade 5 adverse event occurred.

CONCLUSION

Addition of TRT showed significant survival benefits and well tolerability in ES-SCLC patients receiving platinum-etoposide chemotherapy and immune checkpoint inhibitors, which could be a feasible first-line treatment strategy for ES-SCLC patients.

Hybrid Immuno-RT for Bulky Tumors: Standard Fractionation with Partial Tumor SBRT

PURPOSE/OBJECTIVE(S)

Bulky tumors remain challenging to be treated. Stereotactic body radiation therapy (SBRT) is effective against radioresistant tumor cells and can induce immunogenic cell death (ICD) that leads to T-cell-mediated antitumor effects. Low-dose radiation (LDRT) can inflame the tumor microenvironment (TME) by recruiting T cells. We designed a novel radiotherapy technique (RT, ERT) whose dose distribution map resembles the "eclipse" by concurrently delivering LDRT to the whole tumor, meanwhile SBRT to only a part of the same tumor. This study examined the safety and efficacy of ERT to bulky lesions with PD-1 inhibitors in mice and patients.

MATERIALS/METHODS

In mice with CT26 colon or LLC1 lung bulky tumors (400 - 500 cm3), the whole tumor was irradiated by LDRT (2 Gy x 3), meanwhile the tumor center was irradiated by SBRT (10 Gy x 3); αPD-1 was given weekly. The dependence of therapeutic effects on CD8+ T cells was determined using depleting antibodies. Frequencies of CD8+ T cells and M1 macrophages (Mφ) were determined by flow cytometry. Multiplex Immunohistochemistry (mIHC) was applied to analyze the number and the location of CD8+ T cells and their subpopulations, as well as the phospho-eIF2α level (the ICD marker) of tumor cells in TME. Patients with advanced lung or liver bulky tumors who failed standard treatment or with oncologic emergencies were treated. Kaplan-Meier method was applied to estimate patients' progression-free survival (PFS) and overall survival (OS).

RESULTS

ERT/αPD-1 is superior to SBRT/αPD-1 or LDRT/αPD-1 in controlling bulky tumors in both mouse models in a CD8+ T-cell dependent manner. In the CT26 model, ERT/αPD-1 resulted in complete tumor regression in 3/11 mice and induced more CD8+ T cells and M1 Mφ in TME compared to other groups. mIHC analysis showed that ERT/αPD-1 induced higher bulk, stem-like (TCF1+ TIM3- PD-1+), and more differentiated (TCF1- TIM3+ PD-1+) CD8+ T cells infiltration into the tumor center and periphery compared to other groups. Compared to untreated or LDRT-treated tumor centers, tumor centers irradiated with ERT or SBRT showed elevated phospho-eIF2α accompanied by higher dendritic cell infiltration. In total, 39 advanced cancer patients were treated with ERT/αPD-1 or plus chemotherapy. Radiation-induced pneumonitis occurred in 1 of 26 patients receiving thoracic ERT. There were two cases of grade III toxicity associated with PD-1 inhibitors. No toxicity above grade III was observed. The objective response rate was 38.5%. The median PFS was 5.6 months and median OS was not reached at a median follow-up of 11.7 months.

CONCLUSION

ERT/αPD-1 showed superior efficacy in controlling bulky tumor in two mouse models. The hybrid immuno-RT (ERT) combing PD-1 inhibitors was safe and effective in patients with bulky tumors. Further clinical trials in combination with bioimaging to identify the optimal SBRT target region for the bulky tumor are warranted.

DE-JANet: A unified network based on dual encoder and joint attention for Alzheimer's disease classification using multi-modal data

Structural magnetic resonance imaging (sMRI), which can reflect cerebral atrophy, plays an important role in the early detection of Alzheimer's disease (AD). However, the information provided by analyzing only the morphological changes in sMRI is relatively limited, and the assessment of the atrophy degree is subjective. Therefore, it is meaningful to combine sMRI with other clinical information to acquire complementary diagnosis information and achieve a more accurate classification of AD. Nevertheless, how to fuse these multi-modal data effectively is still challenging. In this paper, we propose DE-JANet, a unified AD classification network that integrates image data sMRI with non-image clinical data, such as age and Mini-Mental State Examination (MMSE) score, for more effective multi-modal analysis. DE-JANet consists of three key components: (1) a dual encoder module for extracting low-level features from the image and non-image data according to specific encoding regularity, (2) a joint attention module for fusing multi-modal features, and (3) a token classification module for performing AD-related classification according to the fused multi-modal features. Our DE-JANet is evaluated on the ADNI dataset, with a mean accuracy of 0.9722 and 0.9538 for AD classification and mild cognition impairment (MCI) classification, respectively, which is superior to existing methods and indicates advanced performance on AD-related diagnosis tasks.

Altered Effective Connectivity Among the Cerebellum and Cerebrum in Patients with Major Depressive Disorder Using Multisite Resting-State fMRI

Major depressive disorder (MDD) is a serious and widespread psychiatric disorder. Previous studies mainly focused on cerebrum functional connectivity, and the sample size was relatively small. However, functional connectivity is undirected. And, there is increasing evidence that the cerebellum is also involved in emotion and cognitive processing and makes outstanding contributions to the symptomology and pathology of depression. Therefore, we used a large sample size of resting-state functional magnetic resonance imaging (rs-fMRI) data to investigate the altered effective connectivity (EC) among the cerebellum and other cerebral cortex in patients with MDD. Here, from the perspective of data-driven analysis, we used two different atlases to divide the whole brain into different regions and analyzed the alterations of EC and EC networks in the MDD group compared with healthy controls group (HCs). The results showed that compared with HCs, there were significantly altered EC in the cerebellum-neocortex and cerebellum-basal ganglia circuits in MDD patients, which implied that the cerebellum may be a potential biomarker of depressive disorders. And, the alterations of EC brain networks in MDD patients may provide new insights into the pathophysiological mechanisms of depression.

Research on named entity recognition of Traditional Chinese Medicine chest discomfort cases incorporating domain vocabulary features

OBJECTIVE

To promote research on knowledge extraction and knowledge graph construction of chest discomfort medical cases in Traditional Chinese Medicine (TCM), this paper focuses on their named entity recognition (NER). The recognition task includes six entity types: "syndrome", "symptom", "etiology and pathogenesis", "treatment method", "medication", and "prescription".

METHODS

We annotated data in a BIO (B-begin, I-inside, O-outside) manner. For the characteristics of medical case texts, we proposed a custom dictionary method that can be dynamically updated for word segmentation. To compare the effect of the method on the experimental results, we applied the method in the BiLSTM-CRF model and IDCNN-CRF model, respectively.

RESULTS

The models using custom dictionaries (BiLSTM-CRF-Loaded and IDCNN-CRF-Loaded) outperformed the models without custom dictionaries (BiLSTM-CRF and IDCNN-CRF) in accuracy, precision, recall, and F1 score. The BiLSTM-CRF-Loaded model yielded F1 scores of 92.59% and 93.23% on the test set and validation set, respectively, surpassing the BERT-BiLSTM-CRF model by 3.59% and 4.87%. Furthermore, when analyzing the results for the six entity categories separately, we found that the use of custom dictionaries has a marked impact, with the categories of "etiology and pathogenesis" and "syndrome" demonstrating the most noticeable improvements. By comparing the F1 scores, we observed that the entity category "medication" yielded the highest performance, reaching F1 scores of 96.04% and 96.48% on the test set and validation set, respectively.

CONCLUSION

We propose a word segmentation method based on a dynamically updated custom dictionary. The method is combined with the BILSTM-CRF and the IDCNN-CRF models, which enhances the model to recognize domain-specific terms and new entities. It can be widely applied in dealing with complex text structures and texts containing domain-specific terms.

AC-E Network: Attentive Context-Enhanced Network for Liver Segmentation

Segmentation of liver from CT scans is essential in computer-aided liver disease diagnosis and treatment. However, the 2DCNN ignores the 3D context, and the 3DCNN suffers from numerous learnable parameters and high computational cost. In order to overcome this limitation, we propose an Attentive Context-Enhanced Network (AC-E Network) consisting of 1) an attentive context encoding module (ACEM) that can be integrated into the 2D backbone to extract 3D context without a sharp increase in the number of learnable parameters; 2) a dual segmentation branch including complemental loss making the network attend to both the liver region and boundary so that getting the segmented liver surface with high accuracy. Extensive experiments on the LiTS and the 3D-IRCADb datasets demonstrate that our method outperforms existing approaches and is competitive to the state-of-the-art 2D-3D hybrid method on the equilibrium of the segmentation precision and the number of model parameters.

PA-LBF: Prefix-Based and Adaptive Learned Bloom Filter for Spatial Data

The recently proposed learned bloom filter (LBF) opens a new perspective on how to reconstruct bloom filters with machine learning. However, the LBF has a massive time cost and does not apply to multidimensional spatial data. In this paper, we propose a prefix-based and adaptive learned bloom filter (PA-LBF) for spatial data, which efficiently supports the insertion and deletion. The proposed PA-LBF is divided into three parts: (1) the prefix-based classification. The Z-order space-filling curve is used to extract data, prefix it, and classify it. (2) The adaptive learning process. The multiple independent adaptive sub-LBFs are designed to train the suffixes of data, combined with part 1, to reduce the false positive rate (FPR), query, and learning process time consumption. (3) The backup filter uses CBF. Two kinds of backup CBF are constructed to meet the situation of different insertion and deletion frequencies. Experimental results prove the validity of the theory and show that the PA-LBF reduces the FPR by 84.87%, 79.53%, and 43.01% with the same memory usage compared with the LBF on three real-world spatial datasets. Moreover, the time consumption of PA-LBF can be reduced to $5\times$ and $2.05\times$ that of the LBF on the query and learning process, respectively.

[Characteristics and related factors of viral nucleic acid negative conversion in children infected with Omicron variant strain of SARS-CoV-2]

Objective: To understand the characteristics and associated factors of viral nucleic acid conversion in children infected with Omicron variant strain of SARS-CoV-2 in Shanghai. Methods: The clinical symptoms, laboratory results and other data of 177 children infected with SARS-CoV-2 who were hospitalized in Shanghai Children's Hospital, School of Medicine, Shanghai Jiao Tong University (designated hospital for SARS-CoV-2 infection in Shanghai) from April 25 to June 8, 2022 were retrospectively analyzed. According to the chest imaging findings, the children were divided into mild and common type groups. According to their age, the unvaccinated children were divided into<3 years old group and 3-<18 years old group. According to the vaccination status, the children aged 3-<18 year were divided into non-vaccination group, 1-dose vaccination group and 2-dose vaccination group. Comparison between groups was performed by independent sample t-test and analysis of variance, and multivariate linear regression analysis was used for multivariate analysis. Results: Among the 177 children infected with Omicron variant of SARS-CoV-2, 96 were males and 81 were females, aged 3 (1, 6) years. The time of viral nucleic acid negative conversion was (10.3±3.1) days. The 177 children were 138 cases of mild type and 39 cases of common type. Among the children aged 3-<18 years old, 55 cases were not vaccinated, 5 cases received 1-dose and 36 cases received 2-dose vaccination. Among the 36 children who received 2 doses of vaccination, the time of viral nucleic acid negative conversion was shorter in those vaccinated within 6 months than those over 6 months ((7.1±1.9) vs. (10.8±3.0) d, t=-3.23, P=0.004). Univariate analysis showed that the time of nucleic acid negative conversion of SARS-CoV-2 was associated with age, underlying diseases, gastrointestinal symptoms, white blood cell count, proportion of neutrophils, proportion of lymphocytes, and the number of doses of SARS-CoV-2 vaccine (t=3.87, 2.55, 2.04, 4.24, 3.51, 2.92, F=16.27, all P<0.05). Multiple linear regression analysis showed that older age (β=-0.33, 95% CI -0.485--0.182, P<0.001) and more doses of vaccination (β=-0.79, 95% CI -1.463--0.120, P=0.021) were associated with shortened nucleic acid negative conversion time in children, while lower lymphocyte proportion (β=-0.02, 95% CI -0.044--0.002, P=0.031) and underlying diseases (β=1.52, 95% CI 0.363-2.672, P=0.010) were associated with prolonged nucleic acid negative conversion time in children. Conclusion: The children infected with Omicron variant of SARS-CoV-2 with reduced lymphocyte proportion and underlying diseases may have longer time of viral nucleic acid negative conversion,while children with older age and more doses of vaccination may have shorter time of viral nucleic acid negative conversion.

[The impact of diabetic retinopathy on vision-related quality of life]

Objective: To assess the effect of diabetic retinopathy (DR) on vision-related quality of life (VRQoL) in patients with type 2 diabetes. Methods: In this cross-sectional study, patients with type 2 diabetes residing in 15 residency communities in Fushun, Liaoning province were enrolled from July 2012 to May 2013. We measured the VRQoL by the 25-item National Eye Institute Visual Function Questionnaire (NEI-VFQ-25). Patients were grouped according to their age, gender, presence of visual impairment, and affected eyes. NEI-VFQ-25 scores were compared between/among groups using the Wilcoxon rank-sum test or Kruskal-Wallis H test. The severity of DR in the eyes was graded into no DR, mild non-proliferative diabetic retinopathy (NPDR), moderate NPDR, severe NPDR, and proliferative diabetic retinopathy (PDR). Severity scores from both eyes were then summarized to create a single per-person grade ranging from 1 (no DR in either eye) to 7 (bilateral PDR). Generalized linear models were used to assess the linear relationship between NEI-VFQ-25 scores and DR severity. Locally weighted scatterplot smoothing plots were generated to evaluate the possible nonlinear associations between concatenated severity of DR and VRQoL. Results: A total of 1 537 patients were recruited, including 836 (54.4%) with no DR, 479 (31.2%) with mild NPDR, 90 (5.9%) with moderate NPDR, 72 (4.7%) with severe NPDR and 60 (3.9%) with PDR. Compared with patients with unilateral DR, bilaterally involved subjects were statistically significantly compromised in general vision [70.2 (66.5, 72.5) vs. 68.9 (63.9, 71.6), Z=90.222, P=0.038], near activities [90.5 (85.8, 94.0) vs. 88.8 (84.5, 92.5), Z=114.942, P=0.005], dependency [91.1 (85.6, 96.5) vs. 89.3 (83.8, 94.5), Z=91.934, P=0.033], mental health [80.0 (73.4, 84.9) vs. 77.5 (70.8, 82.0), Z=118.388, P=0.003], role difficulties [76.8 (70.1, 82.4) vs. 74.5 (67.6, 80.6), Z =90.791, P=0.036] and NEI-VFQ-25 composite [88.3 (84.2, 91.0) vs. 86.9 (82.8, 90.1), Z=96.207, P=0.024]. Scores on general vision (χ²=85.665), near activities (χ²=78.462), distance activities (χ²=145.489), social function (χ²=53.629), dependency (χ²=86.710), mental health (χ²=68.281), role difficulties (χ²=45.357), color vision (χ²=68.176), peripheral vision (χ²=116.179) and NEI-VFQ-25 composite (χ²=133.291) decreased gradually as DR severity increased (all P<0.001). On role difficulties, locally weighted scatterplot smoothing plots showed significant"turning points"from bilateral mild NPDR to mild NPDR/>mild NPDR (slope m=-4.7) and from moderate NPDR/≥moderate NPDR to severe NPDR/≥severe NPDR (slope m=-12.6). Conclusion: Both greater severity and bilaterality of DR were associated with lower vision-specific VRQoL, particularly role difficulties and mental health.

An effective automatic segmentation of abdominal adipose tissue using a convolution neural network

BACKGROUND AND AIMS

Computer-aided diagnosis and prognosis rely heavily on fully automatic segmentation of abdominal fat tissue using Emission Tomography images. The identification of subcutaneous adipose tissue (SAT) and visceral adipose tissue (VAT) in abdomen fat faces two main challenges: (1) the great difficulties in comparison to multi-stage semantic segmentation (VAT and SAT), and (2) the subtle differences due to the high similarity of the two classes in abdomen fat and complicated VAT distribution.

METHODS

In this research, we built an automated convolutional neural network (A-CNN) for segmenting Abdominal adipose tissue (AAT) from radiology images.

RESULTS

We developed a point-to-point design for the A-CNN learning process, wherein the representing features might be learned together with a hybrid feature extraction technique. We tested the proposed model on a CT dataset and evaluated it to existing CNN models. Furthermore, our suggested approach, A-CNN, outperformed existing deep learning methods regarding segmentation outcomes, notably in the AAT segment.

CONCLUSIONS

Proposed method is extremely fast with remarkable performance on limited-scale low dose CT-scanning and demonstrates the strength in providing an efficient computer-aimed tool for segmentation of AAT in the clinic.

The alterations of brain functional connectivity networks in major depressive disorder detected by machine learning through multisite rs-fMRI data

BACKGROUND

The current diagnosis of major depressive disorder (MDD) is mainly based on the patient's self-report and clinical symptoms. Machine learning methods are used to identify MDD using resting-state functional magnetic resonance imaging (rs-fMRI) data. However, due to large site differences in multisite rs-fMRI data and the difficulty of sample collection, most of the current machine learning studies use small sample sizes of rs-fMRI datasets to detect the alterations of functional connectivity (FC) or network attribute (NA), which may affect the reliability of the experimental results.

METHODS

Multisite rs-fMRI data were used to increase the size of the sample, and then we extracted the functional connectivity (FC) and network attribute (NA) features from 1611 rs-fMRI data (832 patients with MDD (MDDs) and 779 healthy controls (HCs)). ComBat algorithm was used to harmonize the data variances caused by the multisite effect, and multivariate linear regression was used to remove age and sex covariates. Two-sample t-test and wrapper-based feature selection methods (support vector machine recursive feature elimination with cross-validation (SVM-RFECV) and LightGBM's "feature_importances_" function) were used to select important features. The Shapley additive explanations (SHAP) method was used to assign the contribution of features to the best classification effect model.

RESULTS

The best result was obtained from the LinearSVM model trained with the 136 important features selected by SVMRFE-CV. In the nested five-fold cross-validation (consisting of an outer and an inner loop of five-fold cross-validation) of 1611 data, the model achieved the accuracy, sensitivity, and specificity of 68.90 %, 71.75 %, and 65.84 %, respectively. The 136 important features were tested in a small dataset and obtained excellent classification results after balancing the ratio between patients with depression and HCs.

CONCLUSIONS

The combined use of FC and NA features is effective for classifying MDDs and HCs. The important FC and NA features extracted from the large sample dataset have some generalization performance and may be used as a reference for the altered brain functional connectivity networks in MDD.

A strategy of model space search for dynamic causal modeling in task fMRI data exploratory analysis

Dynamic causal modeling (DCM) is a tool used for effective connectivity (EC) estimation in neuroimage analysis. But it is a model-driven analysis method, and the structure of the EC network needs to be determined in advance based on a large amount of prior knowledge. This characteristic makes it difficult to apply DCM to the exploratory brain network analysis. The exploratory analysis of DCM can be realized from two perspectives: one is to reduce the computational cost of the model; the other is to reduce the model space. From the perspective of model space reduction, a model space exploration strategy is proposed, including two algorithms. One algorithm, named GreedyEC, starts with reducing EC from full model, and the other, named GreedyROI, start with adding EC from one node model. Then the two algorithms were applied to the task state functional magnetic resonance imaging (fMRI) data of visual object recognition and selected the best DCM model from the perspective of model comparison based on Bayesian model compare method. Results show that combining the results of the two algorithms can further improve the effect of DCM exploratory analysis. For convenience in application, the algorithms were encapsulated into MATLAB function based on SPM to help neuroscience researchers to analyze the brain causal information flow network. The strategy provides a model space exploration tool that may obtain the best model from the perspective of model comparison and lower the threshold of DCM analysis.

Deep learning-based automatic tumor burden assessment of pediatric high-grade gliomas, medulloblastomas, and other leptomeningeal seeding tumors

BACKGROUND

Longitudinal measurement of tumor burden with magnetic resonance imaging (MRI) is an essential component of response assessment in pediatric brain tumors. We developed a fully automated pipeline for the segmentation of tumors in pediatric high-grade gliomas, medulloblastomas, and leptomeningeal seeding tumors. We further developed an algorithm for automatic 2D and volumetric size measurement of tumors.

METHODS

The preoperative and postoperative cohorts were randomly split into training and testing sets in a 4:1 ratio. A 3D U-Net neural network was trained to automatically segment the tumor on T1 contrast-enhanced and T2/FLAIR images. The product of the maximum bidimensional diameters according to the RAPNO (Response Assessment in Pediatric Neuro-Oncology) criteria (AutoRAPNO) was determined. Performance was compared to that of 2 expert human raters who performed assessments independently. Volumetric measurements of predicted and expert segmentations were computationally derived and compared.

RESULTS

A total of 794 preoperative MRIs from 794 patients and 1003 postoperative MRIs from 122 patients were included. There was excellent agreement of volumes between preoperative and postoperative predicted and manual segmentations, with intraclass correlation coefficients (ICCs) of 0.912 and 0.960 for the 2 preoperative and 0.947 and 0.896 for the 2 postoperative models. There was high agreement between AutoRAPNO scores on predicted segmentations and manually calculated scores based on manual segmentations (Rater 2 ICC = 0.909; Rater 3 ICC = 0.851). Lastly, the performance of AutoRAPNO was superior in repeatability to that of human raters for MRIs with multiple lesions.

CONCLUSIONS

Our automated deep learning pipeline demonstrates potential utility for response assessment in pediatric brain tumors. The tool should be further validated in prospective studies.

Multi-Label Classification Scheme Based on Local Regression for Retinal Vessel Segmentation

Segmenting small retinal vessels with width less than 2 pixels in fundus images is a challenging task. In this paper, in order to effectively segment the vessels, especially the narrow parts, we propose a local regression scheme to enhance the narrow parts, along with a novel multi-label classification method based on this scheme. We consider five labels for blood vessels and background in particular: the center of big vessels, the edge of big vessels, the center as well as the edge of small vessels, the center of background, and the edge of background. We first determine the multi-label by the local de-regression model according to the vessel pattern from the ground truth images. Then, we train a convolutional neural network (CNN) for multi-label classification. Next, we perform a local regression method to transform the previous multi-label into binary label to better locate small vessels and generate an entire retinal vessel image. Our method is evaluated using two publicly available datasets and compared with several state-of-the-art studies. The experimental results have demonstrated the effectiveness of our method in segmenting retinal vessels.

Altered Effective Connectivity of Children and Young Adults With Unilateral Amblyopia: A Resting-State Functional Magnetic Resonance Imaging Study

The altered functional connectivity (FC) in amblyopia has been investigated by many studies, but the specific causality of brain connectivity needs to be explored further to understand the brain activity of amblyopia. We investigated whether the effective connectivity (EC) of children and young adults with amblyopia was altered. The subjects included 16 children and young adults with left eye amblyopia and 17 healthy controls (HCs). The abnormalities between the left/right primary visual cortex (PVC) and the other brain regions were investigated in a voxel-wise manner using the Granger causality analysis (GCA). According to the EC results in the HCs and the distribution of visual pathways, 12 regions of interest (ROIs) were selected to construct an EC network. The alteration of the EC network of the children and young adults with amblyopia was analyzed. In the voxel-wise manner analysis, amblyopia showed significantly decreased EC between the left/right of the PVC and the left middle frontal gyrus/left inferior frontal gyrus compared with the HCs. In the EC network analysis, compared with the HCs, amblyopia showed significantly decreased EC from the left calcarine fissure, posterior cingulate gyrus, left lingual gyrus, right lingual gyrus, and right fusiform gyrus to the right calcarine fissure. Amblyopia also showed significantly decreased EC from the right inferior frontal gyrus and right lingual gyrus to the left superior temporal gyrus compared with the HCs in the EC network analysis. The results may indicate that amblyopia altered the visual feedforward and feedback pathway, and amblyopia may have a greater relevance with the feedback pathway than the feedforward pathway. Amblyopia may also correlate with the feedforward of the third visual pathway.

Discrepancies between actual weight, weight perception and weight loss intention amongst persons with NAFLD

BACKGROUND

Currently, weight loss remains the main management strategy for NAFLD, but the weight loss intention and methods remain poorly characterized.

METHODS

We analysed data about the perception of weight status, intention and methods to lose weight amongst 3,822 persons with NAFLD (United States Fatty Liver Index ≥ 30) from the National Health and Nutrition Examination Survey, 2001-2014.

RESULTS

Only 53.9% of people with NAFLD intended to lose weight, 91.8% with perception of overweight and 8.2% with normal weight perception. Persons with perception of overweight or overweight/obese status were four times more likely to try to lose weight (adjusted odds ratios 3.9 and 4.2, respectively, both P < 0.0001). Younger age, women, higher educational level, Hispanic and blacks (versus whites) were significant independent factors associated with weight loss intention. Notably, ≤10% attended weight loss programme. Metabolic equivalent of task hours per week was significantly higher in whites who exercised to lose weight (vs. no exercise, P = 0.003) but not in other racial/ethnic groups. Interestingly, calorie intake was similar between those who dieted versus not (2056 vs. 1970 kcal/day, P = 0.11). About 30% reported ≥ 10-lb weight loss, with 50% higher odds of success for men but there was no difference by race/ethnicity.

CONCLUSION

Overweight or obese perception was a key driver in weight loss activities but was inconsistent with actual weight status and varied by race/ethnicity and other sociodemographic factors. Weight loss programme is under-utilized and should take in account of weight perception training and culturally appropriate approach.

Automatic microaneurysm detection in fundus image based on local cross-section transformation and multi-feature fusion

BACKGROUND AND OBJECTIVE

Retinal microaneurysm (MA) is one of the earliest clinical signs of diabetic retinopathy(DR). Its detection is essential for controlling DR and preventing vision loss. However, the spatial scale of MA is extremely small and the contrast to surrounding background is subtle, which make MA detection challenging. The purpose of this work is to automatically detect MAs from fundus images.

METHODS

Our MA detector involves two stages: MA candidate extraction and classification. In MA candidate extraction stage, local minimum region extraction and block filtering are used to exploit the regions where MA may exist. In this way, most of irrelavent background regions are discarded , which subsequently facilitates the training of MA classifier. In the second stage, multiple features are extracted to train the MA classifier. To distinguish MA from vascular regions, we propose a series of descriptors according to the cross-section profile of MA. Specially, as MAs are small and their contrast to surroundings is subtle, we propose local cross-section transformation (LCT) to amplify the difference between the MA and confusing structures. Finally, an under-sampling boosting-based classifier (RUSBoost) is trained to determine whether the candidate is an MA.

RESULTS

The proposed method is evaluated on three public available databases i.e. e-ophtha-MA, DiaretDB1 and ROC training set. It achieves high sensitivities for low false positive rates on the three databases. Using the FROC metric, the final scores are 0.516, 0.402 and 0.293 respectively, which are comparable to existing state-of-the-art methods.

CONCLUSIONS

The proposed local cross-section transformation enhances the discrimination of descriptors by amplifying difference between MAs and confusing structures, which facilitates the classification and improves the detection performances. With the powerful descriptors, our method achieves state-of-the-art performances on three public datasets consistently.