The development and validation of clinical prediction models based on artificial intelligence (AI) and machine learning methods have become increasingly widespread. However, the prediction model bias risk and applicability evaluation tool developed in 2019 (i.e., PROBAST-2019) has shown significant limitations. Therefore, an expanded and updated version of the PROBAST-2019 tool was released in 2025, known as the PROBAST+AI tool. The tool is divided into two parts including model development and model evaluation. It aims to comprehensively and systematically evaluate potential methodological quality issues in model development, bias risks in model evaluation, and the applicability of models, regardless of the modeling method used. This paper provides a systematic interpretation of the PROBAST+AI tool's items and case analyses, with the aim of guiding and assisting researchers engaged in related studies and promoting the high-quality development of clinical predictive model research.

Objective To compare the effects of two head holders on the setup errors during head and neck neoplasms radiotherapy.Methods Totally 30 patients were divided into groups A and B randomly and equally.Group A applied TypeSTM Extension for setup,and group B used Type-S Overlay Board,with thermoplastic film applied to fixation.Varian23IX image guided linear accelerator was involved in the therapy,and the patients underwent CBCT-guided therapy once a week in 6 weeks.The setup errors at left-right,head-foot and anterior-posterior directions were obtained by bone registration and manual checking.Results The absolute setup error at anterior-posterior direction of group A,was (2.31 ± 1.35) mm,which was significantly higher than (1.88±1.79) mm of group B (P=0.032).There were no obvious differences between the setup errors of group A and those of group B (P＞0.05),and the weekly absolute errors of group A were not statistically different from those of group B (P＞0.05).Conclusion There are no significant differences between the effects of the two head holders on the setup errors during head and neck neoplasms radiotherapy,and proper head holder can be selected according to clinical requirements.

Objective To analyze the clinical features of 35 patients with myasthenia gravis (MG) associated with other autoimmune diseases (ADMG). Methods One hundred seventy MG patients were divided into 2 groups. One group included MG patients with autoimmune diseases ( n=35), and the other included MG patients without auto?immune diseases (NADMG, n=135). Clinical features such as gender, age of onset, types of MG, myasthenic crisis, thy?mus abnormalities, different therapies, ocular MG progressing to general MG, and relapse of MG within one year were compared between ADMG and NADMG. Results 85.7%ADMG patients were female which was higher compared with NADMG (P< 0.001); median age of onset was 40 years old (23~45 years old) in ADMG group which was younger than NADMG group. Ocular MG and thymic hyperplasia were more frequent in ADMG group than in NADMG group (45.7%vs. 24.4%, P=0.013;14.3%vs. 1.5%,P=0.001,respectively). Other clinical features such as myasthenic crisis, thymoma, different therapies, ocular MG progressing to general MG, and relapse of MG within one year were not signifi?cantly different between the two groups. In multivariate analysis, female (OR=4.76, 95% CI:1.64-13.77; P= 0.004), age of onset (OR=0.96, 95%CI:0.92-0.99;P=0.019), ocular MG (OR=3.10, 95%CI:1.30-7.41;P=0.011) and thy? mic hyperplasia(OR=16.26, 95%CI:2.22-119.11;P=0.006) were significantly different between the two groups. Con?clusions ADMG is more common in female patients with ocular MG and thymic hyperplasia.