Artificial intelligence in cellular senescence research: a systematic review assessing methodological quality and reporting standards using PROBAST + AI and TRIPOD + AI
Cellular senescence is a fundamental mechanism of biological ageing that has emerged as a critical target for therapeutic intervention in age related diseases. The coalesce of artificial intelligence and senescence research provides unprecedented opportunities in advancing our knowledge and treatment approaches. This systematic review study addresses the gap across diverse AI model and the heterogeneity in senescence, by conducting the extensive evaluation of the performance outcomes and methodological rigor of AI models using Prediction model Risk of Bias Assessment Tool + Artificial Intelligence (PROBAST + AI) and reporting completeness using Transparent Reporting of a multivariable prediction model for Individual Prognosis or Diagnosis + Artificial Intelligence extension (TRIPOD + AI), providing the insights into AI models robustness and generalizability. For the articles, released between 2019 to 2025, across major databases 18 eligible studies was used for review, following PRISMA guideline. Quality and applicability were assessed by PROBAST + AI (4 domain) and reporting via TRIPOD + AI (27 items). The quantitative synthesis indicates that deep learning architectures, especially Convolutional Neural Networks (CNNs), are dominant which appeared in about 50% of the studies. These CNNs consistently outperform traditional machine learning methods in the analysis of morphological heterogeneity. While reported performance metrics were high, with accuracy ranging from 83.55% to 99.79%, the PROBAST + AI assessment indicates a high risk of bias in 83.33% (15/18) of studies, primarily driven by Analysis domain due to improper data splitting (data leakage) and lack of external validation. As well as adherence to TRIPOD + AI reporting standards was suboptimal with average of 62% ‘YES’; notably, with the major gap in 0% of studies pre-registered a protocol and only 44.4% made analytical code publicly available, severely limiting reproducibility. Evidently AI demonstrates immense potential to accelerate biomarker discovery and senolytic drug screening, particularly through label-free morphological analysis by DL models, despite of high quality concern and poor reproducibility limit reliability; also standardization, shared benchmarks, multi-omics integration, and explainable AI are essential concerns for clinical translation in aging research.