Developing an interpretable machine learning model for diagnosing gout using clinical and ultrasound features

European Journal of Radiology, Год журнала: 2025, Номер 184, С. 111959 - 111959

Опубликована: Янв. 31, 2025

Язык: Английский

---

Non-invasive classification of non-neoplastic and neoplastic gallbladder polyps based on clinical imaging and ultrasound radiomics features: An interpretable machine learning model

European Journal of Surgical Oncology, Год журнала: 2025, Номер 51(6), С. 109709 - 109709

Опубликована: Фев. 25, 2025

Язык: Английский

---

Trends and Determinants of Global Infectious Disease Burden from 1990 to 2021: Insights from Machine Learning Models

Опубликована: Янв. 1, 2025

Язык: Английский

---

A Systematic Literature Review of the Latest Advancements in XAI

Technologies, Год журнала: 2025, Номер 13(3), С. 93 - 93

Опубликована: Март 1, 2025

This systematic review details recent advancements in the field of Explainable Artificial Intelligence (XAI) from 2014 to 2024. XAI utilises a wide range frameworks, techniques, and methods used interpret machine learning (ML) black-box models. We aim understand technical future directions. followed PRISMA methodology selected 30 relevant publications three main databases: IEEE Xplore, ACM, ScienceDirect. Through comprehensive thematic analysis, we categorised research into topics: ‘model developments’, ‘evaluation metrics methods’, ‘user-centred system design’. Our results uncover ‘What’, ‘How’, ‘Why’ these were developed. found that 13 papers focused on model developments, 8 studies evaluation metrics, 12 user-centred design. Moreover, it was aimed bridge gap between outputs user understanding.

Язык: Английский

---

Interpretable prognostic modeling for long-term survival of Type A aortic dissection patients using support vector machine algorithm

European journal of medical research, Год журнала: 2025, Номер 30(1)

Опубликована: Апрель 15, 2025

This study aims to develop a reliable and interpretable predictive model for long-term survival in Type A aortic dissection (TAAD) patients, utilizing machine learning (ML) algorithms. We retrospectively reviewed the clinical data of patients diagnosed with TAAD who underwent open surgical repair at First Affiliated Hospital Chongqing Medical University, from September 2017 December 2020, University Central between October 2019 April 2020. Cases less than 20% missing were imputed using random forest To identify significant prognostic factors, we performed LASSO (Least Absolute Shrinkage Selection Operator) Cox regression analysis, including preoperative blood markers, previous medical history intraoperative condition. Based on advantages characteristics set, subsequently developed learning-based Support Vector Machine (SVM) evaluated its performance across key metrics. further explain decision-making process SVM model, employed SHapley Additive exPlanation (SHAP) values interpretation. total 171 included training internal test groups; 73 external group. Through regression, univariate relevance assessment, seven feature variables selected modeling. Performance evaluation revealed that showed excellent both sets, no overfitting, indicating strong applicability. In achieved an AUC 0.9137 (95% CI 0.9081-0.9203) testing 0.8533 0.8503-0.8624) 0.8770 0.8698-0.8982), respectively. The accuracy 0.8366, 0.8481 0.8030; precision 0.8696, 0.8374 0.8235; recall 0.8421, 0.7933 0.7651; F1 scores 0.8290, 0.8148 0.7928; Brier 0.1213, 0.1417 0.1323; average (AP) 0.9019, 0.8789 0.8548, SHAP analysis longer operation time, extended cardiopulmonary bypass (CPB) duration, prolonged cross-clamp (ACC) advanced age, higher plasma transfusion volume, elevated serum creatinine increased white cell (WBC) count significantly contributed predictions. based algorithm assess patients. demonstrated accuracy, precision, robustness identifying high-risk providing evidence clinicians.

Язык: Английский

---

Interpretable machine learning models for predicting clinical pregnancies associated with surgical sperm retrieval from testes of different etiologies: a retrospective study

BMC Urology, Год журнала: 2024, Номер 24(1)

Опубликована: Июль 29, 2024

Abstract Background The relationship between surgical sperm retrieval of different etiologies and clinical pregnancy is unclear. We aimed to develop a robust interpretable machine learning (ML) model for predicting using the SHapley Additive exPlanation (SHAP) association from testes etiologies. Methods A total 345 infertile couples who underwent intracytoplasmic injection (ICSI) treatment with due February 2020 March 2023 at reproductive center were retrospectively analyzed. six models used predict ICSI. After evaluating performance characteristics ML models, Extreme Gradient Boosting (XGBoost) was selected as best model, SHAP utilized interpret XGBoost pregnancies reveal decision-making process model. Results Combining area under receiver operating characteristic curve (AUROC), accuracy, precision, recall, F1 score, brier precision-recall (P-R) (AP), has (AUROC: 0.858, 95% confidence interval (CI): 0.778–0.936, accuracy: 79.71%, score: 0.151). global summary plot values shows that female age most important feature influencing output. showed younger in females, bigger testicular volume (TV), non-tobacco use, higher anti-müllerian hormone (AMH), lower follicle-stimulating (FSH) FSH males, temporary ejaculatory disorders (TED) group, not non-obstructive azoospermia (NOA) group all resulted an increased probability pregnancy. Conclusions predicts associated high reliability, robustness. It can provide counseling decisions patients various

Язык: Английский

---

Associations of systemic inflammation and systemic immune inflammation with serum uric acid concentration and hyperuricemia risk: the mediating effect of body mass index

Frontiers in Endocrinology, Год журнала: 2024, Номер 15

Опубликована: Дек. 9, 2024

With the development of lifestyle, elevated uric acid and hyperuricemia have become important factors affecting human health, but biological mechanism risk are still unclear.

Язык: Английский

---