Drug-induced liver injury(DILI)is one of the major challenges in drug development and clinical practice,and effective prevention and con-trol measures remain lacking.Research has shown that DILI is primarily mediated by immune respons-es.Human leukocyte antigen(HLA)alleles are cur-rently the strongest genetic factors reported to be associated with DILI.Due to the low positive predic-tive value of HLA alleles,preemptive HLA genetic screening has limited clinical utility in preventing DILI.However,its high negative predictive value makes it valuable for DILI diagnosis and causality assessment.In recent years,polymorphisms in im-mune-related genes-such as those involved in anti-gen processing and presentation pathways,T-cell receptors,immunostimulatory molecules,and cyto-kines-have been found to be associated with DILI.Future studies combining these genes with HLA analysis may provide deeper mechanistic insights into DILI and facilitate their translational applica-tion in clinical practice,ultimately improving drug safety.

Recently,diffusion models have emerged as a promising paradigm for molecular design and optimization.However,most diffusion-based molecular generative models focus on modeling 2D graphs or 3D geom-etries,with limited research on molecular sequence diffusion models.The International Union of Pure and Applied Chemistry(IUPAC)names are more akin to chemical natural language than the simplified molecular input line entry system(SMILES)for organic compounds.In this work,we apply an IUPAC-guided conditional diffusion model to facilitate molecular editing from chemical natural language to chemical language(SMILES)and explore whether the pre-trained generative performance of diffusion models can be transferred to chemical natural language.We propose DiffIUPAC,a controllable molecular editing diffusion model that converts IUPAC names to SMILES strings.Evaluation results demonstrate that our model out-performs existing methods and successfully captures the semantic rules of both chemical languages.Chemical space and scaffold analysis show that the model can generate similar compounds with diverse scaffolds within the specified constraints.Additionally,to illustrate the model's applicability in drug design,we conducted case studies in functional group editing,analogue design and linker design.

Recently, diffusion models have emerged as a promising paradigm for molecular design and optimization. However, most diffusion-based molecular generative models focus on modeling 2D graphs or 3D geometries, with limited research on molecular sequence diffusion models. The International Union of Pure and Applied Chemistry (IUPAC) names are more akin to chemical natural language than the Simplified Molecular Input Line Entry System (SMILES) for organic compounds. In this work, we apply an IUPAC-guided conditional diffusion model to facilitate molecular editing from chemical natural language to chemical language (SMILES) and explore whether the pre-trained generative performance of diffusion models can be transferred to chemical natural language. We propose DiffIUPAC, a controllable molecular editing diffusion model that converts IUPAC names to SMILES strings. Evaluation results demonstrate that our model outperforms existing methods and successfully captures the semantic rules of both chemical languages. Chemical space and scaffold analysis show that the model can generate similar compounds with diverse scaffolds within the specified constraints. Additionally, to illustrate the model's applicability in drug design, we conducted case studies in functional group editing, analogue design and linker design.

The advantages of deep learning were introduced when appled to CT imaging,and the present situation of deep learning applied to cardiovascular CT imaging was reviewed in terms of image quality enhancement and de-noising,cardiac structure segmentation and quantitative measurement.The problems of deep learning and challenges encountered during the application to cardiovascular CT imaging were analyzed,and the future development directions included cross-institutional collaborative research,enhanced standardization of the data acquisition process and improved model interpretability.[Chinese Medical Equipment Journal,2025,46(10):97-105]

Objective:To investigate the prognosis of postoperative adjuvant therapy for hepatocellular carcinoma after conversion therapy of combined targeted therapy and immunotherapy followed by sequential hepatectomy.Methods:The retrospective cohort study was conducted. The clinicopathological data of 103 patients with initially unresectable hepatocellular carcinoma (HCC) who were admitted to 11 medical centers in China, including Mengchao Hepatobiliary Hospital of Fujian Medical University et al, from November 2019 to May 2023 were collected. There were 83 males and 20 females, aged (54±12)years. All 103 patients underwent conversion therapy of tyrosine kinase inhibitors (TKIs) and immune checkpoint inhibitors (ICIs) successfully followed by sequential hepatectomy, of which 72 patients undergoing postoperative adjuvant therapy were divided into the adjuvant therapy group, and 31 patients undergoing postoperative follow-up monitoring were divided into the follow-up monitoring group. Observation indicators: (1) follow-up and postoperative condi-tions; (2) analysis of factors influencing recurrence-free survival time of patients; (3) stratified ana-lysis. Comparison of count data between group was conducted using the chi-square test or Fisher exact probability. The R software was used to draw survival curves, and the Log-rank test was used for survival analysis. Univariate and multivariate analyses were conducted using the Cox proportional hazard model. Results:(1) Follow-up and postoperative conditions. All 103 patients were followed up for 21.0(range, 1.9?47.2)months, with the median recurrence-free survival time of 28.7 months and the 1-, 2-, 3-year recurrence-free survival rates of 68.6%, 55.6%, 41.2%. The median overall survival time of 103 patients was unreached, and the 1-, 2-, 3-year overall survival rates were 90.9%, 82.1%, 69.6%, respectively. The median recurrence-free survival time was 33.1 months in patients of the adjuvant therapy group, with the 1-, 2-year recurrence-free survival rates as 77.2%, 61.5%. The median recurrence-free survival time was 11.1 months in patients of the follow-up monitoring group, with the 1-, 2-year recurrence-free survival rates as 46.6%, 40.8%. There was a significant difference in recurrence-free survival between the two groups of patients ( χ2=5.492, P<0.05). (2) Analysis of factors influencing recurrence-free survival time of patients. Results of multivariate analy-sis showed that pathologic complete response and postoperative adjuvant therapy were independent factors influencing recurrence-free survival time of HCC patients undergoing conversion therapy of combined targeted therapy and immunotherapy followed by sequential hepatectomy ( hazard ratio=0.297, 0.492, 95% confidence interval as 0.137?0.647, 0.268?0.903, P<0.05). (3) Stratified analysis. Of the 71 patients with non-pathologic complete response, the median recurrence-free survival time of 48 patients in the adjuvant therapy group was 24.0 months, with the 1-, 2-year recurrence-free survival rates as 67.4%, 48.8%. The median recurrence-free survival time of 23 patients with non-pathological complete response in the follow-up monitoring group was 7.4 months, with the 1-, 2-year recurrence-free survival rates as 35.0%, 26.3%. There was a significant difference in recurrence-free survival between the 48 patients with non-pathologic complete response in the adjuvant therapy group and the 23 patients with non-pathologic complete response in the follow-up monitoring group ( χ2=5.241, P<0.05). Conclusion:For HCC patients with conversion therapy of TKIs and ICIs followed by sequential hepatectomy, postoperative adjuvant therapy, compared to postoperative follow-up monitoring, can prolong the recurrence-free survival time of patients, of whom cases with non-pathologic complete response can benefit from adjuvant therapy.

The advantages of deep learning were introduced when appled to CT imaging,and the present situation of deep learning applied to cardiovascular CT imaging was reviewed in terms of image quality enhancement and de-noising,cardiac structure segmentation and quantitative measurement.The problems of deep learning and challenges encountered during the application to cardiovascular CT imaging were analyzed,and the future development directions included cross-institutional collaborative research,enhanced standardization of the data acquisition process and improved model interpretability.[Chinese Medical Equipment Journal,2025,46(10):97-105]

Drug-induced liver injury(DILI)is one of the major challenges in drug development and clinical practice,and effective prevention and con-trol measures remain lacking.Research has shown that DILI is primarily mediated by immune respons-es.Human leukocyte antigen(HLA)alleles are cur-rently the strongest genetic factors reported to be associated with DILI.Due to the low positive predic-tive value of HLA alleles,preemptive HLA genetic screening has limited clinical utility in preventing DILI.However,its high negative predictive value makes it valuable for DILI diagnosis and causality assessment.In recent years,polymorphisms in im-mune-related genes-such as those involved in anti-gen processing and presentation pathways,T-cell receptors,immunostimulatory molecules,and cyto-kines-have been found to be associated with DILI.Future studies combining these genes with HLA analysis may provide deeper mechanistic insights into DILI and facilitate their translational applica-tion in clinical practice,ultimately improving drug safety.

Objective:To investigate the prognosis of postoperative adjuvant therapy for hepatocellular carcinoma after conversion therapy of combined targeted therapy and immunotherapy followed by sequential hepatectomy.Methods:The retrospective cohort study was conducted. The clinicopathological data of 103 patients with initially unresectable hepatocellular carcinoma (HCC) who were admitted to 11 medical centers in China, including Mengchao Hepatobiliary Hospital of Fujian Medical University et al, from November 2019 to May 2023 were collected. There were 83 males and 20 females, aged (54±12)years. All 103 patients underwent conversion therapy of tyrosine kinase inhibitors (TKIs) and immune checkpoint inhibitors (ICIs) successfully followed by sequential hepatectomy, of which 72 patients undergoing postoperative adjuvant therapy were divided into the adjuvant therapy group, and 31 patients undergoing postoperative follow-up monitoring were divided into the follow-up monitoring group. Observation indicators: (1) follow-up and postoperative condi-tions; (2) analysis of factors influencing recurrence-free survival time of patients; (3) stratified ana-lysis. Comparison of count data between group was conducted using the chi-square test or Fisher exact probability. The R software was used to draw survival curves, and the Log-rank test was used for survival analysis. Univariate and multivariate analyses were conducted using the Cox proportional hazard model. Results:(1) Follow-up and postoperative conditions. All 103 patients were followed up for 21.0(range, 1.9?47.2)months, with the median recurrence-free survival time of 28.7 months and the 1-, 2-, 3-year recurrence-free survival rates of 68.6%, 55.6%, 41.2%. The median overall survival time of 103 patients was unreached, and the 1-, 2-, 3-year overall survival rates were 90.9%, 82.1%, 69.6%, respectively. The median recurrence-free survival time was 33.1 months in patients of the adjuvant therapy group, with the 1-, 2-year recurrence-free survival rates as 77.2%, 61.5%. The median recurrence-free survival time was 11.1 months in patients of the follow-up monitoring group, with the 1-, 2-year recurrence-free survival rates as 46.6%, 40.8%. There was a significant difference in recurrence-free survival between the two groups of patients ( χ2=5.492, P<0.05). (2) Analysis of factors influencing recurrence-free survival time of patients. Results of multivariate analy-sis showed that pathologic complete response and postoperative adjuvant therapy were independent factors influencing recurrence-free survival time of HCC patients undergoing conversion therapy of combined targeted therapy and immunotherapy followed by sequential hepatectomy ( hazard ratio=0.297, 0.492, 95% confidence interval as 0.137?0.647, 0.268?0.903, P<0.05). (3) Stratified analysis. Of the 71 patients with non-pathologic complete response, the median recurrence-free survival time of 48 patients in the adjuvant therapy group was 24.0 months, with the 1-, 2-year recurrence-free survival rates as 67.4%, 48.8%. The median recurrence-free survival time of 23 patients with non-pathological complete response in the follow-up monitoring group was 7.4 months, with the 1-, 2-year recurrence-free survival rates as 35.0%, 26.3%. There was a significant difference in recurrence-free survival between the 48 patients with non-pathologic complete response in the adjuvant therapy group and the 23 patients with non-pathologic complete response in the follow-up monitoring group ( χ2=5.241, P<0.05). Conclusion:For HCC patients with conversion therapy of TKIs and ICIs followed by sequential hepatectomy, postoperative adjuvant therapy, compared to postoperative follow-up monitoring, can prolong the recurrence-free survival time of patients, of whom cases with non-pathologic complete response can benefit from adjuvant therapy.

Background::In vitro fertilization (IVF) has emerged as a transformative solution for infertility. However, achieving favorable live-birth outcomes remains challenging. Current clinical IVF practices in IVF involve the collection of heterogeneous embryo data through diverse methods, including static images and temporal videos. However, traditional embryo selection methods, primarily reliant on visual inspection of morphology, exhibit variability and are contingent on the experience of practitioners. Therefore, an automated system that can evaluate heterogeneous embryo data to predict the final outcomes of live births is highly desirable. Methods::We employed artificial intelligence (AI) for embryo morphological grading, blastocyst embryo selection, aneuploidy prediction, and final live-birth outcome prediction. We developed and validated the AI models using multitask learning for embryo morphological assessment, including pronucleus type on day 1 and the number of blastomeres, asymmetry, and fragmentation of blastomeres on day 3, using 19,201 embryo photographs from 8271 patients. A neural network was trained on embryo and clinical metadata to identify good-quality embryos for implantation on day 3 or day 5, and predict live-birth outcomes. Additionally, a 3D convolutional neural network was trained on 418 time-lapse videos of preimplantation genetic testing (PGT)-based ploidy outcomes for the prediction of aneuploidy and consequent live-birth outcomes.Results::These two approaches enabled us to automatically assess the implantation potential. By combining embryo and maternal metrics in an ensemble AI model, we evaluated live-birth outcomes in a prospective cohort that achieved higher accuracy than experienced embryologists (46.1% vs. 30.7% on day 3, 55.0% vs. 40.7% on day 5). Our results demonstrate the potential for AI-based selection of embryos based on characteristics beyond the observational abilities of human clinicians (area under the curve: 0.769, 95% confidence interval: 0.709–0.820). These findings could potentially provide a noninvasive, high-throughput, and low-cost screening tool to facilitate embryo selection and achieve better outcomes. Conclusions::Our study underscores the AI model’s ability to provide interpretable evidence for clinicians in assisted reproduction, highlighting its potential as a noninvasive, efficient, and cost-effective tool for improved embryo selection and enhanced IVF outcomes. The convergence of cutting-edge technology and reproductive medicine has opened new avenues for addressing infertility challenges and optimizing IVF success rates.

Objective To investigate the composition of symptom clusters in early postoperative breast cancer patients and analyze the sentinel symptoms of each cluster of symptoms. Methods A total of 309 patients who underwent mastectomy were conveniently sampled and surveyed using the Chinese version of the Anderson Symptom Inventory. Principal component analysis and varimax orthogonal rotation were employed to analyze the symptom clusters, and their associations were analyzed using the Apriori algorithm model to identify the sentinel symptoms of each cluster of symptoms. Results Three symptom clusters were identified in early postoperative breast cancer patients: neuro-sleep symptom cluster [fatigue (weakness)-distress-pain-sleepiness-restless sleep], sensory-perception symptom cluster (numbness-forgetfulness-shortness of breath-sadness-dry mouth), and digestive system symptom cluster (nausea-vomiting-loss of appetite). Fatigue was the sentinel symptom of the neuro-sleep symptom cluster, numbness was the sentinel symptom of the sensory-perception symptom cluster, and nausea was the sentinel symptom of the digestive system symptom cluster. Conclusion Early postoperative breast cancer patients experience multiple symptom clusters, with sentinel symptoms existing in each cluster. Healthcare staff should develop intervention measures based on sentinel symptoms to improve the efficiency of symptom management and reduce the degree of symptom distress for patients.