Digital ray: enhancing cataractous fundus images using style transfer generative adversarial networks to improve retinopathy detection.

BACKGROUND/AIMS: The aim of this study was to develop and evaluate digital ray, based on preoperative and postoperative image pairs using style transfer generative adversarial networks (GANs), to enhance cataractous fundus images for improved retinopathy detection. METHODS: For eligible cataract patients, preoperative and postoperative colour fundus photographs (CFP) and ultra-wide field (UWF) images were captured. Then, both the original CycleGAN and a modified CycleGAN (C2ycleGAN) framework were adopted for image generation and quantitatively compared using Frechet Inception Distance (FID) and Kernel Inception Distance (KID). Additionally, CFP and UWF images from another cataract cohort were used to test model performances. Different panels of ophthalmologists evaluated the quality, authenticity and diagnostic efficacy of the generated images. RESULTS: A total of 959 CFP and 1009 UWF image pairs were included in model development. FID and KID indicated that images generated by C2ycleGAN presented significantly improved quality. Based on ophthalmologists' average ratings, the percentages of inadequate-quality images decreased from 32% to 18.8% for CFP, and from 18.7% to 14.7% for UWF. Only 24.8% and 13.8% of generated CFP and UWF images could be recognised as synthetic. The accuracy of retinopathy detection significantly increased from 78% to 91% for CFP and from 91% to 93% for UWF. For retinopathy subtype diagnosis, the accuracies also increased from 87%-94% to 91%-100% for CFP and from 87%-95% to 93%-97% for UWF. CONCLUSION: Digital ray could generate realistic postoperative CFP and UWF images with enhanced quality and accuracy for overall detection and subtype diagnosis of retinopathies, especially for CFP.\ TRIAL REGISTRATION NUMBER: This study was registered with ClinicalTrials.gov (NCT05491798).

Cost-effectiveness and cost-utility of a digital technology-driven hierarchical healthcare screening pattern in China.

Utilization of digital technologies for cataract screening in primary care is a potential solution for addressing the dilemma between the growing aging population and unequally distributed resources. Here, we propose a digital technology-driven hierarchical screening (DH screening) pattern implemented in China to promote the equity and accessibility of healthcare. It consists of home-based mobile artificial intelligence (AI) screening, community-based AI diagnosis, and referral to hospitals. We utilize decision-analytic Markov models to evaluate the cost-effectiveness and cost-utility of different cataract screening strategies (no screening, telescreening, AI screening and DH screening). A simulated cohort of 100,000 individuals from age 50 is built through a total of 30 1-year Markov cycles. The primary outcomes are incremental cost-effectiveness ratio and incremental cost-utility ratio. The results show that DH screening dominates no screening, telescreening and AI screening in urban and rural China. Annual DH screening emerges as the most economically effective strategy with 341 (338 to 344) and 1326 (1312 to 1340) years of blindness avoided compared with telescreening, and 37 (35 to 39) and 140 (131 to 148) years compared with AI screening in urban and rural settings, respectively. The findings remain robust across all sensitivity analyses conducted. Here, we report that DH screening is cost-effective in urban and rural China, and the annual screening proves to be the most cost-effective option, providing an economic rationale for policymakers promoting public eye health in low- and middle-income countries.

Assessing evidence-based practice competencies of undergraduate nursing students using a developed multi-criteria decision-analysis model.

AIM: The developed multi-criteria decision analysis model was used to identify the inter-influence relationships and key factors affecting the EBP competencies of UNSs, to assess the EBP competencies of UNSs and based on these results, to formulate an improvement strategy to enhance the EBP competencies of UNSs. BACKGROUND: EBP is considered a core competency in international nursing practice. However, few studies have developed EBP evaluation models and applied them to assessing and improving the EBP competencies of UNSs. DESIGN: This is a quantitative study with multi-criteria decision-analysis model. METHODS: Firstly, the questionnaire was designed based on the characteristics of the DEMATEL and VIKOR-AS methods, which was completed by 17 nursing experts from a case hospital in Zhejiang Province, China. Subsequently, the DEMATEL method was used to analyze the inter-influence relationships among various criteria to determine their respective weights. Finally, the VIKOR method is utilized to integrate multiple criteria and their relative weights to assign comprehensive scores to each UNSs. RESULTS: The use of the DEMATEL method reveals that "Knowledge (C1)", "Mastering the basic scientific research methods during the study of the undergraduate courses (C11)", "Being able to consult clinical experts appropriately when encountering problems in clinical practice (C23)" and "Understanding the importance of reading journals related to the nursing profession regularly (C34)" were critical influencing factors. "Skill (C2)," "Being able to explain the essential roles of the best research evidence in determining clinical practice (C15)," "Being able to apply the collected research evidence to the individual case in nursing care (C25)" and "Paying attention to using the evidence-based nursing practice concept to determine the best clinical practice (C35)" were the most influential factors. According to the VIKOR method, the performance of the UNSs in the case hospitals in terms of EBP competencies from highest to lowest was Student C, Student B and Student A. However, all of these students suffered from deficiencies at the knowledge level. CONCLUSIONS: The application of the DEMATEL and VIKOR methods provides a systematic and comprehensive approach to the assessment of EBP competencies of UNSs. The lack of EBP competencies of UNSs in case hospitals is mainly reflected in knowledge level. To improve UNSs' EBP competencies, medical schools and hospital educators should propose short- and long-term strategies to improve knowledge.

Effect of Premedication With Pronase Before Upper Gastrointestinal Endoscopy: A Multicenter Prospective Randomized Controlled Study.

OBJECTIVES: This study aimed to confirm whether premedication with pronase before endoscopy improves mucosal visualization and increases precancerous lesion and cancer lesion detection rates. MATERIALS AND METHODS: From June 2018 to April 2019, out-patients scheduled for endoscopy from 13 hospitals were screened to be randomly allocated in a 2:1 ratio to premedication with pronase (group A) and water (group B). The primary endpoint was mucosal visibility scores, and the secondary endpoint was precancerous and cancer lesion detection rates. This trial was registered at Chinese Clinical Trial Registry, and the registration number was ChiCTR1800016853. RESULTS: Group A showed significantly lower mucosal visibility scores (better mucosal visibility) of esophagus, stomach, and duodenum than group B, with all P -values <0.001. The overall cancer detection rates between group A and group B were 0.83 and 1.08%, and overall detection rates of precancerous and cancer lesion were 4.4 and 4.9%, both without significant difference ( P =1.000 and 0.824). In addition, the flushing volume (milliliter) of group A (10.52±23.41) was less than group B (36.30±52.11) ( P <0.001), and the flushing frequency of group A (0.46±1.01) was fewer than group B (1.62±2.12) ( P <0.001). CONCLUSIONS: Premedication with pronase could achieve better mucosal visibility and decrease flushing frequency and volume, but may not increase lesion detection rates.

Real-Time Evaluation of Helicobacter pylori Infection by Convolution Neural Network During White-Light Endoscopy: A Prospective, Multicenter Study (With Video).

INTRODUCTION: Convolutional neural network during endoscopy may facilitate evaluation of Helicobacter pylori infection without obtaining gastric biopsies. The aim of the study was to evaluate the diagnosis accuracy of a computer-aided decision support system for H. pylori infection (CADSS-HP) based on convolutional neural network under white-light endoscopy. METHODS: Archived video recordings of upper endoscopy with white-light examinations performed at Sir Run Run Shaw Hospital (January 2019-September 2020) were used to develop CADSS-HP. Patients receiving endoscopy were prospectively enrolled (August 2021-August 2022) from 3 centers to calculate the diagnostic property. Accuracy of CADSS-HP for H. pylori infection was also compared with endoscopic impression, urea breath test (URT), and histopathology. H. pylori infection was defined by positive test on histopathology and/or URT. RESULTS: Video recordings of 599 patients who received endoscopy were used to develop CADSS-HP. Subsequently, 456 patients participated in the prospective evaluation including 189 (41.4%) with H. pylori infection. With a threshold of 0.5, CADSS-HP achieved an area under the curve of 0.95 (95% confidence interval [CI], 0.93-0.97) with sensitivity and specificity of 91.5% (95% CI 86.4%-94.9%) and 88.8% (95% CI 84.2%-92.2%), respectively. CADSS-HP demonstrated higher sensitivity (91.5% vs 78.3%; mean difference = 13.2%, 95% CI 5.7%-20.7%) and accuracy (89.9% vs 83.8%, mean difference = 6.1%, 95% CI 1.6%-10.7%) compared with endoscopic diagnosis by endoscopists. Sensitivity of CADSS-HP in diagnosing H. pylori was comparable with URT (91.5% vs 95.2%; mean difference = 3.7%, 95% CI -1.8% to 9.4%), better than histopathology (91.5% vs 82.0%; mean difference = 9.5%, 95% CI 2.3%-16.8%). DISCUSSION: CADSS-HP achieved high sensitivity in the diagnosis of H. pylori infection in the real-time test, outperforming endoscopic diagnosis by endoscopists and comparable with URT. Clinicaltrials.gov ; ChiCTR2000030724.

FFMAVP: a new classifier based on feature fusion and multitask learning for identifying antiviral peptides and their subclasses.

Antiviral peptides (AVPs) are widely found in animals and plants, with high specificity and strong sensitivity to drug-resistant viruses. However, due to the great heterogeneity of different viruses, most of the AVPs have specific antiviral activities. Therefore, it is necessary to identify the specific activities of AVPs on virus types. Most existing studies only identify AVPs, with only a few studies identifying subclasses by training multiple binary classifiers. We develop a two-stage prediction tool named FFMAVP that can simultaneously predict AVPs and their subclasses. In the first stage, we identify whether a peptide is AVP or not. In the second stage, we predict the six virus families and eight species specifically targeted by AVPs based on two multiclass tasks. Specifically, the feature extraction module in the two-stage task of FFMAVP adopts the same neural network structure, in which one branch extracts features based on amino acid feature descriptors and the other branch extracts sequence features. Then, the two types of features are fused for the following task. Considering the correlation between the two tasks of the second stage, a multitask learning model is constructed to improve the effectiveness of the two multiclass tasks. In addition, to improve the effectiveness of the second stage, the network parameters trained through the first-stage data are used to initialize the network parameters in the second stage. As a demonstration, the cross-validation results, independent test results and visualization results show that FFMAVP achieves great advantages in both stages.

Efficacy and safety of modified tetracycline dosing in a quadruple therapy for Helicobacter pylori: A retrospective single center study.

BACKGROUND: Although highly effective as a component of Helicobacter pylori (H. pylori) treatment regimen, tetracycline is associated with a high incidence of medication-related adverse events. Modified dosing of tetracycline as part of quadruple therapy may improve safety while providing comparable eradication rates. AIM: To evaluate the efficacy and safety of modified dosing of tetracycline in patients receiving tetracycline and furazolidone-containing quadruple therapy in patients with H. pylori infection. METHODS: Consecutive patients (10/2020-12/2021) who received tetracycline and furazolidone quadruple therapy for H. pylori infection at Sir Run Run Shaw Hospital were identified. All patients received tetracycline, furazolidone, proton pump inhibitor, and bismuth for 14 d as primary or rescue therapy. Modified tetracycline dose group received tetracycline 500 mg twice daily while standard group received 750 mg twice daily or 500 mg three times daily. RESULTS: Three hundred and ninety-four patients [mean age = 46.3 ± 13.9, male = 137 (34.8%), and 309 (78.4%) primary therapy] completed tetracycline and furazolidone quadruple therapy for H. pylori infection including those who received modified tetracycline dose in 157 and standard doses in 118 (750 mg twice daily) and 119 (500 mg three times daily). Eradication rates in the modified tetracycline dose group were 92.40% and in the standard groups, eradication rates were 93.20% for 750 mg twice daily group and 92.43% for 500 mg three times daily group, respectively, without statistical difference (P = 0.959). The incidence of adverse events was lower in the modified tetracycline dose (15.3% vs 32.3% and 29.4%; P = 0.002) compared to the standard dose group. CONCLUSION: In a real-world experience, modified tetracycline dosing as part of tetracycline and furazolidone quadruple therapy for 14 d demonstrated high efficacy, comparable to standard tetracycline dose regimens, with a favorable safety profile.

Advances of endoscopic and surgical management in gastrointestinal stromal tumors.

As one of the most common mesenchymal malignancies in the digestive system, gastrointestinal stromal tumors (GISTs) occur throughout the alimentary tract with diversified oncological characteristics. With the advent of the tyrosine kinase inhibitor era, the treatment regimens of patients with GISTs have been revolutionized and GISTs have become the paradigm of multidisciplinary therapy. However, surgery resection remains recognized as the potentially curative management for the radical resection and provided with favorable oncological outcomes. The existing available surgery algorithms in clinical practice primarily incorporate open procedure, and endoscopic and laparoscopic surgery together with combined operation techniques. The performance of various surgery methods often refers to the consideration of risk evaluation of recurrence and metastases; the degree of disease progression; size, location, and growth pattern of tumor; general conditions of selected patients; and indications and safety profile of various techniques. In the present review, we summarize the fundamental principle of surgery of GISTs based on risk assessment as well as tumor size, location, and degree of progress with an emphasis on the indications, strengths, and limitations of current surgery techniques.

Identification of critical quality factors and critical satisfaction gaps in emergency training courses for new nurses by the multi-criteria network structure model.

OBJECTIVE: To identify critical quality factors and critical satisfaction gaps in emergency training courses for new nurses through a systematic decision-making model. METHODS: Firstly, the service quality (SERVQUAL) was used in the evaluation index system of this study. Then, the decision-making trial and evaluation laboratory (DEMATEL) method was used to analyze the relationship structure and the corresponding weights between the indicators. Finally, the importance-performance analysis (IPA) method was used to identify the categories of all indicators and the corresponding strategic directions. Fifteen new nurses in Taizhou Hospital of Zhejiang Province were selected as participants in this study. RESULTS: The IPA results showed that "(C13)," "(C22)," "(C52)," "(C53)," "(C54)," "(C55)," "(C56),"and "(C57)" are critical satisfaction gaps. From the results of influence network and weight, empathy (C5) was the critical quality factor of the entire training course. The influence network relationship structure and weight had a 98.1% significant confidence level, indicating good stability. CONCLUSION: Teachers' empathy is key to the learning outcomes of new nurses in emergency nursing training courses. Hence, teachers should be attentive to the empathetic quality of their teaching methods to help new nurses gain knowledge and experience in emergency care, especially when they come from different professions and departments.

Disease Burden of Total and Early-Onset Colorectal Cancer in China from 1990 to 2019 and Predictions of Cancer Incidence and Mortality.

Background: Understanding the temporal trends in the epidemiology of colorectal cancer (CRC) and early-onset CRC (EOCRC) in China is essential for policymakers to develop appropriate strategies to reduce the CRC burden. Methods: The prevalence, incidence, mortality, years of life lived with disability (YLDs), years of life lost (YLLs), and disability-adjusted life years (DALYs) of CRC were obtained from the Global Burden of Disease (GBD) Study 2019. The incidence and mortality of CRC over the next 25 years were predicted. Results: From 1990 to 2019, the prevalence, incidence, and mortality of total CRC and EOCRC significantly increased in males, with milder trends in females. In 2019, the number of people living with CRC (or EOCRC) in China was approximately 3.4 (0.59) million, which was over seven (five) times higher than that in 1990. The DALYs, YLDs, and YLLs moderately increased from 1990 to 2019 in both sexes. The age-standardized mortality rate (ASMR) for females has shown a stable trend in total CRC, and a downward trend in EOCRC since 2000. While the ASMR for males showed increasing trends in total CRC and EOCRC. In 2019, the highest incidence, prevalence, YLDs, YLLs, and DALYs were all observed in the 65 to 69 age group, while the highest mortality was in the 70 to 74. By 2044, the incidence and deaths of CRC are expected to reach 1310 thousand and 484 thousand, respectively. For EOCRC, the incidence will peak at about 101 thousand around 2034, and the mortality will continuously decrease to a nadir at about 18 thousand around 2044. Conclusion: Although the age-standardized incidence and mortality of total CRC and EOCRC in China will reach a plateau, the number of incident cases and deaths of CRC have been increasing in the last three decades and will continue to increase in the next 25 years.

Effectiveness and safety of vonoprazan-based regimens compared with those of proton pump inhibitor (PPI)-based regimens as first-line agents for Helicobacter pylori: a meta-analysis of randomized clinical trials.

PURPOSE: Vonoprazan (VPZ), a reversible H+-K+ ATPase inhibitor, has a relatively fast and sustained acid-suppression action that is unaffected by diet or gene polymorphisms. Several randomized controlled trials have evaluated the difference in the eradication rate of Helicobacter pylori (HP) between VPZ-based and proton pump inhibitor (PPI)-based regimens. The present review aimed to (1) evaluate the efficacy, safety, and compliance of VPZ-based regimens compared with those of PPI-based regimens as first-line treatments for HP infection and (2) perform a subgroup analysis to examine the influence of differences in clarithromycin-resistance status, treatment duration, treatment regimens, and research region on treatment outcomes. METHODS: We conducted a systematic literature search on PubMed, Embase, Cochrane Library, Web of Science, and ChiCTR Register. Systematic searches, study selection, data extraction, risk of bias assessment, and statistical analysis were performed according to pre-registered protocol on the PROSPERO (CRD42022336608). RESULTS: Eight studies and 2956 HP-infected patients were enrolled. Only first-line therapy and RCT study were considered. VPZ-based group had a superior eradication efficacy compared to PPI-based group by intention-to-treat (ITT) (pooled risk ratio (RR): 1.14, 95% CI: 1.08-1.21, p < 0.00001) and per-protocol analysis (pooled RR: 1.13, 95% CI: 1.07-1.20, p < 0.00001). This finding was further validated by subgroup analysis depending on treatment regimens, duration, region, and clarithromycin resistance. In addition, there was no significant difference in adverse events (p = 0.33) and compliances (p = 0.30) between the regimens. CONCLUSION: The VPZ-based regimens showed a superior eradication efficacy compared to the already frequently used PPI-based regimens. Furthermore, VPZ-based therapy showed comparable tolerability and incidence of adverse events.

SHP2 inhibition mitigates adaptive resistance to MEK inhibitors in KRAS-mutant gastric cancer through the suppression of KSR1 activity.

Despite the promising antitumor activity of RAF/MEK inhibitors for RAS-driven cancers, not all patients respond to these therapies. Adaptive resistance has been reported as a major culprit in non-responders, which can be reversed by SHP2 inhibitors (SHP2is) in multiple cancer cells; however, the underlying mechanisms remain unknown. In this study, we found that KRAS-mutant gastric cancer cells respond to MEK inhibitors (MEKis) with adaptive resistance. Markedly, SHP2 activation accompanied by ERK signaling restoration in MEKi-treated cells, and a MEKi and SHP2i combination had a synergistic effect on downstream signaling blockade. In vivo, SHP099 combined with AZD6244 (selumetinib) was highly efficacious for the treatment of xenografts. Mechanistically, SHP2 was found to interact with the scaffold protein KSR1 through its protein tyrosine phosphatase domain. KSR1 knockdown sensitized cells to AZD6244, whereas a KSR1 activating mutation (S269A) diminished the synergistic anti-proliferative effect of SHP2i and MEKi. Interestingly, activated SHP2, during adaptive resistance to MEKis, impaired the interaction with KSR1, activating KSR1 to promote MAPK signaling. In conclusion, SHP2 promotes adaptive resistance to MEKis by activating KSR1; selumetinib combined with SHP099 might be an available therapeutic strategy for KRAS-mutant gastric cancers.

Mental Health Among University Students, Using the 12-item General Health Questionnaire.

Context: To date, researchers have found that poor mental health was common during the COVID-19 epidemic. Even if they had been relatively resistant to suicidal ideation during the first three waves of the pandemic, university students may experience a delayed impact on their mental health. Objective: The study intended to measure mental health among university students in Wuhu City, China and to identify an effective approach to universities can use to prevent mental-health issues. Design: The research team performed a cross-sectional study. Setting: The study took place at Anhui polytechnic university, Wuhu, China. Participants: Participants were 2371 students at Anhui polytechnic university in Wuhu city, China. Outcome Measures: The research team used the two-item General Health Questionnaire-12 (GHQ-12) to measure participants' mental health. Results: Among the 2371 participants, 1727 had poor mental health (72.84%), including 843 males (48.81%) and 884 females (51.19%). Poor mental health was significantly associated with an urban residential location (P > .01), the female gender (p>0.01), the second school year (P > .01), and the parents' education level of junior high school or below (both P > .01). Conclusions: The current study suggests that poor mental health among university students is common. Being female, from an urban area, and in the second year of school and having parents with an education of junior high school or below had poorer mental health than those who were male, from the countryside, and in the first year of school and who had parents with a higher level of education.

Multiple Bismuth Quadruple Therapy Containing Tetracyclines Combined with Other Antibiotics and Helicobacter pylori Eradication Therapy.

Helicobacter pylori (HP) infection is closely associated with the development of chronic gastritis, peptic ulcer, and gastric cancer. However, the resistance rate of H. pylori strains to antibiotics such as clarithromycin, metronidazole, and levofloxacin has increased significantly, resulting in a significant decrease in the eradication efficacy of commonly used regimens. Tetracycline has received the attention of domestic and foreign scholars because of its low resistance. The purpose of this review is to provide an update on the tetracycline-containing bismuth quadruple eradication therapy for H. pylori infection and review the efficacy and safety of the regimens, hoping to provide guidance for clinical practice.

An end-to-end tracking method for polyp detectors in colonoscopy videos.

Deep learning based computer-aided diagnosis technology demonstrates an encouraging performance in aspect of polyp lesion detection on reducing the miss rate of polyps during colonoscopies. However, to date, few studies have been conducted for tracking polyps that have been detected in colonoscopy videos, which is an essential and intuitive issue in clinical intelligent video analysis task (e.g. lesion counting, lesion retrieval, report generation). In the paradigm of conventional tracking-by-detection system, detection task for lesion localization is separated from the tracking task for cropped lesions re-identification. In the multi object tracking problem, each target is supposed to be tracked by invoking a tracker after the detector, which introduces multiple inferences and leads to external resource and time consumption. To tackle these problems, we proposed a plug-in module named instance tracking head (ITH) for synchronous polyp detection and tracking, which can be simply inserted into object detection frameworks. It embeds a feature-based polyp tracking procedure into the detector frameworks to achieve multi-task model training. ITH and detection head share the model backbone for low level feature extraction, and then low level feature flows into the separate branches for task-driven model training. For feature maps from the same receptive field, the region of interest head assigns these features to the detection head and the ITH, respectively, and outputs the object category, bounding box coordinates, and instance feature embedding simultaneously for each specific polyp target. We also proposed a method based on similarity metric learning. The method makes full use of the prior boxes in the object detector to provide richer and denser instance training pairs, to improve the performance of the model evaluation on the tracking task. Compared with advanced tracking-by-detection paradigm methods, detectors with proposed ITH can obtain comparative tracking performance but approximate 30% faster speed. Optimized model based on Scaled-YOLOv4 detector with ITH illustrates good trade-off between detection (mAP 91.70%) and tracking (MOTA 92.50% and Rank-1 Acc 88.31%) task at the frame rate of 66 FPS. The proposed structure demonstrates the potential to aid clinicians in real-time detection with online tracking or offline retargeting of polyp instances during colonoscopies.

An overview of agents and treatments for PDGFRA-mutated gastrointestinal stromal tumors.

Platelet-derived growth factor receptor A (PDGFRA) mutations occur in approximately 10-15% of gastrointestinal stromal tumors (GISTs). These tumors with PDGFRA mutations have a different pathogenesis, clinical characteristics, and treatment response compared to tumors with receptor tyrosine kinase protein (KIT) mutations (60-70%). Many clinical studies have investigated the use of tyrosine kinase inhibitors mainly in patients with KIT mutations; however, there is a lack of attention to the PDGFRA-mutated molecular subtype. The main effective inhibitors of PDGFRA are ripretinib, avapritinib, and crenolanib, and their mechanisms and efficacy in GIST (as confirmed in clinical trials) are described in this review. Some multi-targeted tyrosine kinase inhibitors with inhibitory effects on this molecular subtype are also introduced and summarized in this paper. This review focuses on PDGFRA-mutated GISTs, introduces their clinical characteristics, downstream molecular signaling pathways, and existing resistance mechanisms. We focus on the most recent literature that describes the development of PDGFRA inhibitors and their use in clinical trials, as well as the potential benefits from different combination therapy strategies.

Might Patients with Metastatic Gastrointestinal Stromal Tumors Benefit from Operative Management? A Population-Based Retrospective Study.

Background: With respect to effect of surgery on the therapy of patients with metastatic gastrointestinal stromal tumors (mGISTs), still no consensus has been reached. This research designed to investigate the effect of surgical treatment on prognosis in patients with mGISTs. Methods: The population-based study consisted of 6282 GIST patients diagnosed between 2001 and 2016, from the Surveillance, Epidemiology, and End Results (SEER) database registry. The Kaplan-Meier method and Cox model were employed for the exploration of the effect of surgery on overall survival (OS) and GIST-specific survival (GSS). Results: In total, 6282 patients were diagnosed with GISTs, including 1238 (19.7%) mGIST patients and 5044 (80.3%) non-mGIST patients. Compared with the patients with non-mGISTs, metastatic patients assumed relatively lower proportion of surgical management (756 [61.1%] vs. 4666 [92.5%], P < 0.001). Based on unadjusted analysis, mGIST patients with operative management presented higher five years OS together with GSS in comparison with those without operative management (OS: 58.3% vs. 33.1%, P < 0.001; GSS: 61.6% vs. 36.7%, P < 0.001). Multivariable analysis found that no surgery was correlated to more than 2-fold increased death risk (OS, adjusted HR = 2.27, 95% CI: 1.90-2.71; GSS, adjusted HR = 2.42, 95% CI: 2.00-2.93). Conclusion: Metastatic GIST patients could potentially benefit from operative management with improved GSS and OS.

Multi-label recognition of cancer-related lesions with clinical priors on white-light endoscopy.

Deep learning-based computer-aided diagnosis techniques have demonstrated encouraging performance in endoscopic lesion identification and detection, and have reduced the rate of missed and false detections of disease during endoscopy. However, the interpretability of the model-based results has not been adequately addressed by existing methods. This phenomenon is directly manifested by a significant bias in the representation of feature localization. Good recognition models experience severe feature localization errors, particularly for lesions with subtle morphological features, and such unsatisfactory performance hinders the clinical deployment of models. To effectively alleviate this problem, we proposed a solution to optimize the localization bias in feature representations of cancer-related recognition models that is difficult to accurately label and identify in clinical practice. Optimization was performed in the training phase of the model through the proposed data augmentation method and auxiliary loss function based on clinical priors. The data augmentation method, called partial jigsaw, can "break" the spatial structure of lesion-independent image blocks and enrich the data feature space to decouple the interference of background features on the space and focus on fine-grained lesion features. The annotation-based auxiliary loss function used class activation maps for sample distribution correction and led the model to present localization representation converging on the gold standard annotation of visualization maps. The results show that with the improvement of our method, the precision of model recognition reached an average of 92.79%, an F1-score of 92.61%, and accuracy of 95.56% based on a dataset constructed from 23 hospitals. In addition, we quantified the evaluation representation of visualization feature maps. The improved model yielded significant offset correction results for visualized feature maps compared with the baseline model. The average visualization-weighted positive coverage improved from 51.85% to 83.76%. The proposed approach did not change the deployment capability and inference speed of the original model and can be incorporated into any state-of-the-art neural network. It also shows the potential to provide more accurate localization inference results and assist in clinical examinations during endoscopies.

Broadening the Mutation Spectrum in GJA8 and CHMP4B: Novel Missense Variants and the Associated Phenotypes in Six Chinese Han Congenital Cataracts Families.

Purpose: To broaden the mutation and phenotype spectrum of the GJA8 and CHMP4B genes and to reveal genotype-phenotype correlations in a cohort of Chinese patients with congenital cataracts (CCs). Methods: Six Chinese Han families with CCs inherited in an autosomal dominant (AD) pattern were recruited for this study. All patients underwent full ocular examinations. Genomic DNA was extracted from the leukocytes of peripheral blood collected from all available patients and their unaffected family members. Whole-exome sequencing (WES) was performed on all probands and at least one of their parents. Candidate variants were further confirmed by Sanger sequencing. Bioinformatic analysis with several computational predictive programs was performed to assess the impacts of the candidate variants on the structure and function of the proteins. Results: Four heterozygous candidate variants in three different genes (CRYBB2, GJA8, and CHMP4B) were identified in affected individuals from the six families, including two novel missense variants (GJA8: c.64G > C/p. G22R, and CHMP4B: c.587C > G/p. S196C), one missense mutation (CRYBB2: c.562C > T/p. R188C), and one small deletion (GJA8: c.426_440delGCTGGAGGGGACCCT/p.143_147delLEGTL). The three missense mutations were predicted as deleterious in all four computational prediction programs. In the homologous model, the GJA8: p.143_147delLEGTL mutation showed a sequence deletion of five amino acids at the cytoplasmic loop of the Cx50 protein, close to the third transmembrane domain. Patients carrying mutations in the same gene showed similar cataract phenotypes at a young age, including total cataracts, Y-sutural with fetal nuclear cataracts, and subcapsular cataracts. Conclusion: This study further expands the mutation spectrum and genotype-phenotype correlation of CRYBB2, GJA8, and CHMP4B underlying CCs. This study sheds light on the importance of comparing congenital cataract phenotypes in patients at the same age stage. It offers clues for the pathogenesis of CCs and allows for an early prenatal diagnosis for families carrying these genetic variants.

Application of Surgical Decision Model for Patients With Childhood Cataract: A Study Based on Real World Data.

Reliable validated methods are necessary to verify the performance of diagnosis and therapy-assisted models in clinical practice. However, some validated results have research bias and may not reflect the results of real-world application. In addition, the conduct of clinical trials has executive risks for the indeterminate effectiveness of models and it is challenging to finish validated clinical trials of rare diseases. Real world data (RWD) can probably solve this problem. In our study, we collected RWD from 251 patients with a rare disease, childhood cataract (CC) and conducted a retrospective study to validate the CC surgical decision model. The consistency of the real surgical type and recommended surgical type was 94.16%. In the cataract extraction (CE) group, the model recommended the same surgical type for 84.48% of eyes, but the model advised conducting cataract extraction and primary intraocular lens implantation (CE + IOL) surgery in 15.52% of eyes, which was different from the real-world choices. In the CE + IOL group, the model recommended the same surgical type for 100% of eyes. The real-recommended matched rates were 94.22% in the eyes of bilateral patients and 90.38% in the eyes of unilateral patients. Our study is the first to apply RWD to complete a retrospective study evaluating a clinical model, and the results indicate the availability and feasibility of applying RWD in model validation and serve guidance for intelligent model evaluation for rare diseases.