{"doi":"10.1101/2024.09.19.24313909","title":"<i>RiskPath</i> : Explainable deep learning for multistep biomedical prediction in longitudinal data","abstract":"Highlights Many diseases are increasingly conceptualized as multifactorial, progressive processes Robust prediction of progressive disease courses can advance risk stratification and treatment targeting RiskPath provides optimizable timeseries AI to predict progressive disease with longitudinal cohort data Enhanced explainability and functionality facilitates risk pathway mapping and compact models The Bigger Picture Identifying persons at elevated risk for a disease outcome is a key prerequisite for targeting interventions to improve health. Current risk stratification tools for common diseases are aging and achieve only moderate performance. Moreover, many diseases are increasingly recognized to be complex outcomes where individual risk is determined not by a single effect modifier but by time-dependent interactions among many contributory factors over the lifecourse. There is an urgent need to improve individual-level prediction for progressive diseases and understand how multifactorial risks interact over time so that risk stratification and accompanying prevention and intervention strategies can be targeted earlier and more effectively in the disease course. Summary Many diseases are the end outcomes of multifactorial risks that interact and increment over months or years. Timeseries AI methods have attracted increasing interest given their ability to operate on native timeseries data to predict disease outcomes. Instantiating such models in risk stratification tools has proceeded more slowly, in part limited by factors such as structural complexity, model size and explainability. Here, we present RiskPath, an explainable AI toolbox that offers advanced timeseries methods and additional functionality relevant to risk stratification use cases in classic and emerging longitudinal cohorts. Theoretically-informed optimization is integrated in prediction to specify optimal model topology or explore performance-complexity tradeoffs. Accompanying modules allow the user to map the changing importance of predictors over the disease course, visualize the most important antecedent time epochs contributing to disease risk or remove predictors to construct compact models for clinical applications with minimal performance impact.","journal":"medRxiv","year":2024,"id":491911,"datarank":0.0,"base_score":0.0,"endowment":0.0,"self_citation_contribution":0.0,"citation_network_contribution":0.0,"self_endowment_contribution":0.0,"citer_contribution":0.0,"corpus_percentile":null,"corpus_rank":null,"citation_count":1,"citer_count":0,"citers_with_citation_signal":0,"citers_with_endowment":0,"datacite_reuse_total":0,"is_dataset":false,"is_dataset_confidence":0.9375,"is_data_producer":false,"deposit_databanks":null,"is_oa":true,"file_count":0,"downloads":0,"has_version_chain":false,"published_date":"2024-01-01","fair_score":null,"fair_percentile":null,"algorithm_id":"datarank_citation_only_1hop_v6","ranking_scope":"data_only","authors":[{"id":1169810,"name":"Michael J. Ramshaw","orcid":"0000-0001-8800-2817","position":1,"is_corresponding":false},{"id":1338229,"name":"Wai Yin Lam","orcid":null,"position":2,"is_corresponding":false},{"id":341708,"name":"Nina de Lacy","orcid":"0000-0002-4574-3604","position":0,"is_corresponding":true}],"reference_count":88,"raw_metadata":null,"created_at":"2026-07-19T02:08:49.768795Z","pmid":"39371168","pmcid":null,"fwci":null,"citation_percentile":null,"influential_citations":0,"oa_status":null,"license":null,"views":0,"total_file_size_bytes":0,"version_count":0,"fair_f":null,"fair_a":null,"fair_i":null,"fair_r":null,"fair_zscore":null,"fair_rationale":null,"fair_model":null,"fair_agent_version":null,"fair_fulltext_source":null,"fair_has_llm":null,"fair_computed_at":null,"clinical_trials":[],"software_tools":[],"db_accessions":[],"linked_datasets":[],"topics":[]}