Learning the natural history of human disease with generative transformers
Decision-making in healthcare relies on understanding patients’ past and current health states to predict and, ultimately, change their future course1-3. Artificial intelligence (AI) methods promise to aid this task by learning patterns of disease progression from large corpora of health records4,5....
Gespeichert in:
| Hauptverfasser: | , , , , , , , |
|---|---|
| Dokumenttyp: | Article (Journal) |
| Sprache: | Englisch |
| Veröffentlicht: |
17 September 2025
|
| In: |
Nature
Year: 2025, Jahrgang: 647, Heft: 8088, Pages: 248-256, [20] |
| ISSN: | 1476-4687 |
| DOI: | 10.1038/s41586-025-09529-3 |
| Online-Zugang: | Verlag, kostenfrei, Volltext: https://doi.org/10.1038/s41586-025-09529-3 Verlag, kostenfrei, Volltext: https://www.nature.com/articles/s41586-025-09529-3 
|
| Verfasserangaben: | Artem Shmatko, Alexander Wolfgang Jung, Kumar Gaurav, Søren Brunak, Laust Hvas Mortensen, Ewan Birney, Tom Fitzgerald & Moritz Gerstung |
MARC
| LEADER | 00000naa a2200000 c 4500 | ||
|---|---|---|---|
| 001 | 1950813088 | ||
| 003 | DE-627 | ||
| 005 | 20260202105104.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 260202s2025 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.1038/s41586-025-09529-3 |2 doi | |
| 035 | |a (DE-627)1950813088 | ||
| 035 | |a (DE-599)KXP1950813088 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rda | ||
| 041 | |a eng | ||
| 084 | |a 33 |2 sdnb | ||
| 100 | 1 | |a Shmatko, Artem |e VerfasserIn |0 (DE-588)1277649278 |0 (DE-627)183061276X |4 aut | |
| 245 | 1 | 0 | |a Learning the natural history of human disease with generative transformers |c Artem Shmatko, Alexander Wolfgang Jung, Kumar Gaurav, Søren Brunak, Laust Hvas Mortensen, Ewan Birney, Tom Fitzgerald & Moritz Gerstung |
| 264 | 1 | |c 17 September 2025 | |
| 300 | |b Diagramme | ||
| 300 | |a 29 | ||
| 336 | |a Text |b txt |2 rdacontent | ||
| 337 | |a Computermedien |b c |2 rdamedia | ||
| 338 | |a Online-Ressource |b cr |2 rdacarrier | ||
| 500 | |a Enthält 20 weitere ungezählte Seiten | ||
| 500 | |a Gesehen am 02.02.2026 | ||
| 520 | |a Decision-making in healthcare relies on understanding patients’ past and current health states to predict and, ultimately, change their future course1-3. Artificial intelligence (AI) methods promise to aid this task by learning patterns of disease progression from large corpora of health records4,5. However, their potential has not been fully investigated at scale. Here we modify the GPT6 (generative pretrained transformer) architecture to model the progression and competing nature of human diseases. We train this model, Delphi-2M, on data from 0.4 million UK Biobank participants and validate it using external data from 1.9 million Danish individuals with no change in parameters. Delphi-2M predicts the rates of more than 1,000 diseases, conditional on each individual’s past disease history, with accuracy comparable to that of existing single-disease models. Delphi-2M’s generative nature also enables sampling of synthetic future health trajectories, providing meaningful estimates of potential disease burden for up to 20 years, and enabling the training of AI models that have never seen actual data. Explainable AI methods7 provide insights into Delphi-2M’s predictions, revealing clusters of co-morbidities within and across disease chapters and their time-dependent consequences on future health, but also highlight biases learnt from training data. In summary, transformer-based models appear to be well suited for predictive and generative health-related tasks, are applicable to population-scale datasets and provide insights into temporal dependencies between disease events, potentially improving the understanding of personalized health risks and informing precision medicine approaches. | ||
| 650 | 4 | |a Computational science | |
| 650 | 4 | |a Diseases | |
| 650 | 4 | |a Risk factors | |
| 700 | 1 | |a Jung, Alexander Wolfgang |e VerfasserIn |4 aut | |
| 700 | 1 | |a Gaurav, Kumar |e VerfasserIn |4 aut | |
| 700 | 1 | |a Brunak, Søren |e VerfasserIn |4 aut | |
| 700 | 1 | |a Mortensen, Laust Hvas |e VerfasserIn |4 aut | |
| 700 | 1 | |a Birney, Ewan |e VerfasserIn |4 aut | |
| 700 | 1 | |a Fitzgerald, Tom |e VerfasserIn |4 aut | |
| 700 | 1 | |a Gerstung, Moritz |e VerfasserIn |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |t Nature |d London [u.a.] : Nature Publ. Group, 1869 |g 647(2025), 8088, Seite 248-256, [20] |h Online-Ressource |w (DE-627)240151402 |w (DE-600)1413423-8 |w (DE-576)079718426 |x 1476-4687 |7 nnas |a Learning the natural history of human disease with generative transformers |
| 773 | 1 | 8 | |g volume:647 |g year:2025 |g number:8088 |g pages:248-256, [20] |g extent:29 |a Learning the natural history of human disease with generative transformers |
| 856 | 4 | 0 | |u https://doi.org/10.1038/s41586-025-09529-3 |x Verlag |x Resolving-System |z kostenfrei |3 Volltext |7 0 |
| 856 | 4 | 0 | |u https://www.nature.com/articles/s41586-025-09529-3 |x Verlag |z kostenfrei |3 Volltext |7 0 |
| 951 | |a AR | ||
| 992 | |a 20260202 | ||
| 993 | |a Article | ||
| 994 | |a 2025 | ||
| 998 | |g 1277649278 |a Shmatko, Artem |m 1277649278:Shmatko, Artem |d 140000 |e 140000PS1277649278 |k 0/140000/ |p 1 |x j | ||
| 999 | |a KXP-PPN1950813088 |e 4872382846 | ||
| BIB | |a Y | ||
| SER | |a newspaper | ||
| JSO | |a {"origin":[{"dateIssuedKey":"2025","dateIssuedDisp":"17 September 2025"}],"note":["Enthält 20 weitere ungezählte Seiten","Gesehen am 02.02.2026"],"recId":"1950813088","name":{"displayForm":["Artem Shmatko, Alexander Wolfgang Jung, Kumar Gaurav, Søren Brunak, Laust Hvas Mortensen, Ewan Birney, Tom Fitzgerald & Moritz Gerstung"]},"relHost":[{"pubHistory":["1.1869/70 - ; auch mit durchgehender Nr.-Zählung"],"part":{"volume":"647","extent":"29","year":"2025","pages":"248-256, [20]","issue":"8088","text":"647(2025), 8088, Seite 248-256, [20]"},"type":{"media":"Online-Ressource","bibl":"newspaper"},"id":{"zdb":["1413423-8"],"issn":["1476-4687"],"eki":["240151402"]},"origin":[{"publisher":"Nature Publ. Group ; Macmillan","dateIssuedDisp":"1869-","publisherPlace":"London [u.a.] ; London","dateIssuedKey":"1869"}],"titleAlt":[{"title":"Nature <London>"},{"title":"Nature"}],"recId":"240151402","title":[{"title":"Nature","subtitle":"international weekly journal of science","title_sort":"Nature"}],"physDesc":[{"extent":"Online-Ressource"}],"language":["eng"],"disp":"Learning the natural history of human disease with generative transformersNature","note":["Gesehen am 19.02.2019"]}],"title":[{"title_sort":"Learning the natural history of human disease with generative transformers","title":"Learning the natural history of human disease with generative transformers"}],"id":{"eki":["1950813088"],"doi":["10.1038/s41586-025-09529-3"]},"type":{"bibl":"article-newspaper","media":"Online-Ressource"},"physDesc":[{"extent":"29 S.","noteIll":"Diagramme"}],"language":["eng"],"person":[{"family":"Shmatko","given":"Artem","display":"Shmatko, Artem","role":"aut"},{"given":"Alexander Wolfgang","family":"Jung","role":"aut","display":"Jung, Alexander Wolfgang"},{"family":"Gaurav","given":"Kumar","display":"Gaurav, Kumar","role":"aut"},{"role":"aut","display":"Brunak, Søren","given":"Søren","family":"Brunak"},{"display":"Mortensen, Laust Hvas","role":"aut","family":"Mortensen","given":"Laust Hvas"},{"family":"Birney","given":"Ewan","display":"Birney, Ewan","role":"aut"},{"display":"Fitzgerald, Tom","role":"aut","family":"Fitzgerald","given":"Tom"},{"given":"Moritz","family":"Gerstung","role":"aut","display":"Gerstung, Moritz"}]} | ||
| SRT | |a SHMATKOARTLEARNINGTH1720 | ||