Metabolic dysfunction-associated steatotic liver disease (MASLD) comprises a spectrum of liver injuries, including steatosis to steatohepatitis (MASH), liver fibrosis, cirrhosis, and relevant complications. The liver mainly comprises hepatocytes, liver sinusoidal endothelial cells (LSECs), Kupffer cells (KCs), immune cells (T cells, B cells), and hepatic stellate cells (HSCs). Crosstalk among these different liver cells, endogenous aberrant glycolipid metabolism, and altered gut dysbiosis are involved in the pathophysiology of MASLD. This review systematically examines advances in understanding the molecular pathogenesis of MASLD, with a focus on emerging therapeutic targets and translational clinical trials. We first delineate the crucial regulatory mechanisms involving diverse liver cells and the gut-liver axis in MASLD development. These cell-specific pathogenic insights offer valuable perspectives for advancing precision medicine approaches in MASLD treatment. Furthermore, we evaluate potential therapeutic targets and summarize clinical trials currently underway. By comprehensively updating the MASLD pathophysiology and identifying promising strategies, this review aims to facilitate the development of novel pharmacotherapies for this increasingly prevalent condition.
Humans ; Fatty Liver/therapy* ; Animals ; Liver/pathology* ; Kupffer Cells/metabolism* ; Hepatocytes/metabolism* ; Hepatic Stellate Cells/metabolism*

Objective To explore the predictive value of fat attenuation index(FAI)and fibrous plaque index(FPI)for the prognosis of patients with acute coronary syndrome(ACS).Methods A retrospective cohort study was conducted on 334 ACS patients undergoing percutaneous coronary intervention(PCI)in the First Affiliated Hospital of Army Military Medical University and Yongchuan Hospital of Chongqing Medical University from March 2021 to July 2023.All patients received coronary computed tomography angiography(CCTA)to measure FAI and FPI.According to the occurrence of major adverse cardiovascular events(MACE)with 1 year of follow-up,they were divided into MACE group(n=108)and non-MACE group(n=226).The baseline data,CCTA data and results of laboratory tests were collected and compared between the 2 groups.Multivariate logistic regression analysis was used to analyze the relationship of FAI and FPI with the prognosis of ACS patients,and ROC curve was drawn to evaluate its predictive efficiency.Results Among the 334 ACS patients,108(32.34%)experienced MACE.When compared with the non-MACE group,the MACE group exhibited significantly larger proportions of diabetes(72.22%vs 31.86%)and left main coronary artery disease(18.52%vs 7.08%),but lower success rate of operation(79.63%vs 93.81%,P<0.05).Radiologic results showed that the proportion of severe stenosis(20.37%vs 10.62%),FAI(-80.12±6.41 HU vs-72.34±7.09 HU)and FPI(0.58±0.41 vs 0.26±0.12)were obviously increased in the MACE group than the non-MACE group(P<0.05).Laboratory tests indicated that there were statistical differences between the 2 groups in high-density lipoprotein-cholesterol(HDL-C,1.20±0.15 vs 1.09±0.16 mmol/L),miR-126(0.91±0.12 vs 0.96±0.15)and SST2(38.45±5.67 vs 34.30±4.89 ng/mL,P<0.05).Multivariate Logistic regression analysis revealed that FAI(OR=1.200,95%CI:1.136～1.268),FPI(OR=63.157,95%CI:14.126～282.374),moderate stenosis(OR=1.332,95%CI:1.024～1.859),severe stenosis(OR=1.480,95%CI:1.074～2.039),miR-126(OR=0.007,95%CI:0.001～0.077),and sST2(OR=1.192,95%CI:1.113～1.277)were independent predictors of MACE(P<0.05).ROC curve analysis displayed that stenosis degree(AUC=0.622,95%CI:0.561～0.683,P=0.001),FAI(AUC=0.790,95%CI:0.741～0.839,P=0.001)and FPI(AUC=0.700,95%CI:0.638～0.761,P=0.001),miR-126(AUC=0.646,95%CI:0.584～0.707,P=0.001),sST2(AUC=0.700,95%CI:0.638～0.761,P=0.001)had certain predictive values for ACS prognosis.Conclusion Coronary FAI and FPI can be used as independent prognostic indicators of ACS patients,and their numerical changes are closely related to plaque stability and inflammatory state.

Purpose To analyze the value of a deep learning(DL)radiomics model based on dynamic contrast-enhanced MRI images in predicting the expression of Ki-67 status in breast cancer.Materials and Methods A retrospective analysis of 152 breast cancer patients confirmed by pathological results at Guangdong Women and Children Hospital,MRI images and clinical pathological data were reviewed,and based on postoperative immunohistochemistry results,the images of the high and low expression groups of Ki-67 were randomly sampled in a ratio of 8∶2 to form a training set of 122 cases and a validation set of 30 cases.Single-factor and multi-factor Logistic regression analyses of clinical data were performed to select independent predictors of breast cancer expressing Ki-67 status.The ResNet-18 model was used as the basic model for DL feature extraction.Hand-crafted radiomic features and DL features were extracted.Eight machine learning models were constructed based on clinical features,hand-crafted radiomic features,DL features,and their combinations.The area under the receiver operating characteristic curve was used to evaluate the predictive performance of the models,and the best model was determined as the output model.Results The progesterone receptor status(OR=0.764,P=0.040)and human epidermal growth factor receptor-2 status(OR=1.187,P=0.046)were independent clinical predictors of breast cancer expressing Ki-67 status.The combined feature models demonstrated superior performance over the individual feature models,and the support vector machine algorithm had the highest prediction performance in the validation set,with an area under the curve of 0.847.Conclusion The DL radiomics model based on dynamic contrast-enhanced MRI images can effectively predict the expression of Ki-67 status in breast cancer.The support vector machine algorithm combined with feature model is the best,which can help the clinical diagnosis and treatment of breast cancer and prognosis evaluation.

Objective:To describe the hotspots and application trends of artificial intelligence (AI) in epidemiology in the past decade and analyze its advantages and challenges.Methods:The literatures with AI and epidemiology related keywords were systematically retrieved from Web of Science and China National Knowledge Infrastructure from 2014 to 2024. CiteSpace was used for bibliometric analysis of publication volume, keyword co-occurrence, clustering, emergence and cited literature co-occurrence analysis.Results:A total of 5 389 English papers and 1 659 Chinese papers were included, showing an increasing publication trend. High-frequency Chinese keywords included prediction, influencing factor, and machine learning, while English keywords frequently used were machine learning, prediction, and artificial intelligence. The Chinese keywords formed 14 clusters such as epidemiological characteristic, dietary pattern, and elderly individual, and the English keywords formed 21 clusters including prediction model, risk factor, and adult. In international studies, health policy, COVID-19, and digital health were the emerging frontier keywords. Eleven core papers were selected, covering key areas like traffic accident risk assessment, public health big data application, and deep learning in medical diagnosis.Conclusions:This study systematically summarized the research hotspots and development trends of AI applications in epidemiology over the past decade by using bibliometric methods, which indicated that current AI-based epidemiological studies are still in the exploratory phase, with the coexisting of both advantages and challenges. Continued attention should be paid to the future development of this field.

Objective:To investigate the clinical efficacy of combination of a Flow-through anterolateral thigh myocutaneous flap with a contralateral great saphenous vein grafting in reconstruction of Gustilo ⅢC foot and ankle injury in emergency surgery.Methods:This retrospective study analysed the clinical data of 9 patients who were admitted in the Department of Wound Repair Surgery, Suzhou Ruihua Orthopaedic Hospital between September 2021 and October 2023. All patients were diagnosed with Gustilo ⅢC foot and ankle injury and met the predefined inclusion criteria. The average age of the patients was 45 (26-64) years. The mean of Mangled Extremity Severity Score (MESS) was 8.2 points, with a range from 5 to 10 points. The wound area after debridement ranged from 15.5 cm×6.0 cm to 24.5 cm×12.5 cm, with the vascular occlusion or defect at 4.0-20.0 (mean, 6.3) cm in length. The size of the flap was 16.5 cm×7.0 cm-25.5 cm×13.5 cm. During surgery, segments of the great saphenous vein were excised in the length of 5.0-21.0 cm, at an approximately 6.5 cm in length. The Flow-through anterolateral thigh myocutaneous flap was employed to reconstruct the soft tissue defect and to establish vascular connections between the anterior or posterior tibial arteries and veins, with the vessels carried by the flap. Additionally, the contralateral great saphenous vein was taken to bridge the posterior or anterior tibial arteries and veins. Follow-ups were carried out by outpatient visits, telephone calls and WeChat interviews. The flap viability, limb blood circulation, wound healing at both donor and recipient sites, as well as functional recovery of the affected limbs was observed. Sensation recovery of the flap was assessed according to the rating standard established by the British Medical Research Council (BMRC) , while the function recovery of foot and ankle was evaluated using the American Orthopaedic Foot and Ankle Society (AOFAS) scale.Results:All flaps successfully survived without vascular complication. Eight affected limbs were effectively cured, while a calf that had an infection and necrosis of a long segment of tibia and fibula was amputated below-the-knee in the stage-II surgery. One of the donor sites of flap was covered with a medium-thickness skin graft from lower abdomen due to the large size of the flap, while the other donor sites were directly sutured and healed at first intention. One patient developed postoperative bone infection, which was successfully treated through re-debridement and antibiotic bone cement filling. Another patient experienced partial necrosis at the distal hallux due to a prolonged limb ischemia, and a stump repair surgery was performed. All patients were entered into the postoperative follow-up for 15.3 (6-24) months. At the final follow-up visit, satisfactory colour and texture were observed on all flaps, although minor oedema was present. Donor sites healed well and the donor limb function was not affected. Sensory evaluation rated S 2 in 5 flaps and S 3 in 3 flaps according to BMRC. Functional assessment of foot and ankle of the affected limb yielded excellent in 2 patients, good in 5 patients and fair in 1 patient according to AOFAS. Conclusion:Flow-through anterolateral thigh myocutaneous flap combined with a contralateral great saphenous vein grafting enables one-stage reconstruction for soft tissue defect and blood circulation of the affected limb. It is a good method in the treatment of Gustilo ⅢC foot and ankle injury in emergency surgery.

Objective:To explore the clinical effects of free bilateral turbocharged anterolateral thigh flaps in tandem in repairing extensive wounds in the foot and ankle.Methods:The study was a retrospective observational study. From April 2020 to June 2023, 12 patients with extensive wounds in the foot and ankle who met the inclusion criteria were admitted to the Department of Wound Repair Surgery of Suzhou Ruihua Orthopedic Hospital, including 8 males and 4 females, aged 21 to 65 years. The wound area after debridement ranged from 27 cm×14 cm to 37 cm×20 cm. The bilateral perforator flaps pedicled with either oblique or descending branches of the lateral circumflex femoral artery were designed and harvested based on the size and shape of the wounds. The individual flap incision area ranged from 16 cm×9 cm to 34 cm×12 cm. The non-homologous perforator of the flap on the one side was turbocharged by anastomosing it with the gross muscular branch or main vessel of the oblique or descending branch of the lateral circumflex femoral artery from the flap. Subsequently, the proximal end of the oblique or descending branch of the lateral circumflex femoral artery and its accompanying vein from the flap on the one side were connected end-to-end with either the anterior tibial artery and vein, posterior tibial artery and vein, or dorsal foot artery and vein in the recipient area, the distal end of the oblique or descending branch of the lateral circumflex femoral artery and its accompanying vein from the flap on the one side were anastomosed end-to-end with a source vessel originating from flap on the other side. The wounds in the flap donor areas were sutured directly. The number and source of perforators carried by the flaps and the duration of the flap repair surgery were recorded. The survival of the flap, the occurrence of vascular crisis, and the wound healing at both donor and recipient areas were observed after surgery. The flap condition, appearance and function of the affected limb were observed during follow-up. At the last follow-up, the sensory function of the flap was assessed using the British Medical Research Council's sensory rating standard, the foot and ankle function of the affected limb was evaluated according to the American Orthopedic Foot and Ankle Society scoring standard.Results:A total 24 flaps were successfully harvested, carrying 60 perforators, including 34 perforators from the oblique branch of the lateral circumflex femoral artery, 24 perforators from the descending branch of the lateral circumflex femoral artery, one perforator from the transverse branch of the lateral circumflex femoral artery, and one perforator from the direct branch of the femoral artery. The duration of the flap repair surgery ranged from 4.2 to 9.0 hours. The flaps of 12 patients exhibited complete survival after surgery. A total of two flaps of two patients experienced venous crisis after surgery but survived through emergency exploration. One patient encountered undesirable wound healing at the donor area of flap on the one side after surgery, which healed after dressing change, debridement, and suturing. The remaining patients' donor area wounds healed. Two patients displayed impaired wound healing in the recipient area, which improved after dressing change and resection of residual sequestrum, and the wounds in the recipient area of other patients healed successfully. During the follow-up of 4-26 months, the flaps demonstrated favorable color and texture, slight edematous appearance, and partial sensory recovery, as well as good aesthetic and functional restoration of the affected limbs. At the last follow-up, the sensory function of the flap was assessed as grade S2 in 9 cases and grade S3 in 3 cases; the foot and ankle function of the affected limb was evaluated as excellent in two cases, good in 9 cases, and fair in one case.Conclusions:The bilateral turbocharged anterolateral thigh flaps have numerous sources of perforators. By implementing supercharging of non-homologous perforators within the flap, the vascular supply to the flap is turbocharged, thereby mitigating the risk of extensive flap necrosis. The flap is an effective approach for repairing extensive wounds in the foot and ankle, resulting in improved function of the affected limb after repair.

Objective To observe the efficacy of Venovo venous stent for treating post-thrombotic syndrome(PTS)after lower extremity deep venous thrombosis(DVT).Methods A total of 61 patients with PTS after lower extremioty DVT who underwent iliac vein stent implantation were retrospectively enrolled,including 33 cases underwent Venovo stent(Venovo stent group)and 28 cases underwent conventional stent implantation(conventional stent group).The technical success rates and perioperative complications,Villalta score and venous clinical severity score(VCSS)before and 6 and 12 months after treatment,as well as the primary stent patency were compared within and between groups.Results The technical success rate was 100%in both groups,and no significant difference of the incidence of perioperative complications was found between groups(P=0.187).There was no significant difference of Villalta score nor VCSS between groups before treatment(both P＞0.05),while significant decrease of Villalta score was noticed before and 6 or 12 months after treatment within both groups(all P＜0.05).No significant difference of Villalta score nor VCSS was found 6 months after treatment between groups(both P＞0.05),while significant difference of Villalta score and VCSS were observed in Venovo stent group compared with those in conventional stent group 12 months after treatment(both P＜0.05).No significant difference of primary stent patency rate was found between groups 6 and 12 months after treatment(both P＞0.05).Conclusion Venovo venous stent could be used to effectively treat PTS after lower extremity DVT.

Objective:To describe the hotspots and application trends of artificial intelligence (AI) in epidemiology in the past decade and analyze its advantages and challenges.Methods:The literatures with AI and epidemiology related keywords were systematically retrieved from Web of Science and China National Knowledge Infrastructure from 2014 to 2024. CiteSpace was used for bibliometric analysis of publication volume, keyword co-occurrence, clustering, emergence and cited literature co-occurrence analysis.Results:A total of 5 389 English papers and 1 659 Chinese papers were included, showing an increasing publication trend. High-frequency Chinese keywords included prediction, influencing factor, and machine learning, while English keywords frequently used were machine learning, prediction, and artificial intelligence. The Chinese keywords formed 14 clusters such as epidemiological characteristic, dietary pattern, and elderly individual, and the English keywords formed 21 clusters including prediction model, risk factor, and adult. In international studies, health policy, COVID-19, and digital health were the emerging frontier keywords. Eleven core papers were selected, covering key areas like traffic accident risk assessment, public health big data application, and deep learning in medical diagnosis.Conclusions:This study systematically summarized the research hotspots and development trends of AI applications in epidemiology over the past decade by using bibliometric methods, which indicated that current AI-based epidemiological studies are still in the exploratory phase, with the coexisting of both advantages and challenges. Continued attention should be paid to the future development of this field.

Purpose To analyze the value of a deep learning(DL)radiomics model based on dynamic contrast-enhanced MRI images in predicting the expression of Ki-67 status in breast cancer.Materials and Methods A retrospective analysis of 152 breast cancer patients confirmed by pathological results at Guangdong Women and Children Hospital,MRI images and clinical pathological data were reviewed,and based on postoperative immunohistochemistry results,the images of the high and low expression groups of Ki-67 were randomly sampled in a ratio of 8∶2 to form a training set of 122 cases and a validation set of 30 cases.Single-factor and multi-factor Logistic regression analyses of clinical data were performed to select independent predictors of breast cancer expressing Ki-67 status.The ResNet-18 model was used as the basic model for DL feature extraction.Hand-crafted radiomic features and DL features were extracted.Eight machine learning models were constructed based on clinical features,hand-crafted radiomic features,DL features,and their combinations.The area under the receiver operating characteristic curve was used to evaluate the predictive performance of the models,and the best model was determined as the output model.Results The progesterone receptor status(OR=0.764,P=0.040)and human epidermal growth factor receptor-2 status(OR=1.187,P=0.046)were independent clinical predictors of breast cancer expressing Ki-67 status.The combined feature models demonstrated superior performance over the individual feature models,and the support vector machine algorithm had the highest prediction performance in the validation set,with an area under the curve of 0.847.Conclusion The DL radiomics model based on dynamic contrast-enhanced MRI images can effectively predict the expression of Ki-67 status in breast cancer.The support vector machine algorithm combined with feature model is the best,which can help the clinical diagnosis and treatment of breast cancer and prognosis evaluation.

Objective To observe the efficacy of Venovo venous stent for treating post-thrombotic syndrome(PTS)after lower extremity deep venous thrombosis(DVT).Methods A total of 61 patients with PTS after lower extremioty DVT who underwent iliac vein stent implantation were retrospectively enrolled,including 33 cases underwent Venovo stent(Venovo stent group)and 28 cases underwent conventional stent implantation(conventional stent group).The technical success rates and perioperative complications,Villalta score and venous clinical severity score(VCSS)before and 6 and 12 months after treatment,as well as the primary stent patency were compared within and between groups.Results The technical success rate was 100%in both groups,and no significant difference of the incidence of perioperative complications was found between groups(P=0.187).There was no significant difference of Villalta score nor VCSS between groups before treatment(both P＞0.05),while significant decrease of Villalta score was noticed before and 6 or 12 months after treatment within both groups(all P＜0.05).No significant difference of Villalta score nor VCSS was found 6 months after treatment between groups(both P＞0.05),while significant difference of Villalta score and VCSS were observed in Venovo stent group compared with those in conventional stent group 12 months after treatment(both P＜0.05).No significant difference of primary stent patency rate was found between groups 6 and 12 months after treatment(both P＞0.05).Conclusion Venovo venous stent could be used to effectively treat PTS after lower extremity DVT.