Nikita Makarov
@nikitamakarov.bsky.social
LLMs & Digital Twins for Cancer | PhD student at Roche pRED & Helmholtz Munich | Opinions are my own
Pinned
DT-GPT: showing that LLMs can forecast patient trajectories (1/10)
Now in npj Digital Medicine www.nature.com/articles/s41...
Also in Doctor Penguin!
Big thanks to Maria Bordukova, @raulrod.bsky.social Papichaya Quengdaeng Daniel Garger @fschmich.bsky.social Michael Menden Helmholtz Munich Roche
Now in npj Digital Medicine www.nature.com/articles/s41...
Also in Doctor Penguin!
Big thanks to Maria Bordukova, @raulrod.bsky.social Papichaya Quengdaeng Daniel Garger @fschmich.bsky.social Michael Menden Helmholtz Munich Roche
Paper here: www.nature.com/articles/s41...
“Large Language Models forecast Patient Health Trajectories enabling Digital Twins”
“Large Language Models forecast Patient Health Trajectories enabling Digital Twins”
Large language models forecast patient health trajectories enabling digital twins - npj Digital Medicine
npj Digital Medicine - Large language models forecast patient health trajectories enabling digital twins
www.nature.com
October 7, 2025 at 7:39 AM
Paper here: www.nature.com/articles/s41...
“Large Language Models forecast Patient Health Trajectories enabling Digital Twins”
“Large Language Models forecast Patient Health Trajectories enabling Digital Twins”
Overall, DT-GPT shows that LLMs have the potential to become human digital twins. We hope that, in the future, LLM based digital twins will revolutionize the way we run clinical trials & patient care (10/10).
October 7, 2025 at 7:39 AM
Overall, DT-GPT shows that LLMs have the potential to become human digital twins. We hope that, in the future, LLM based digital twins will revolutionize the way we run clinical trials & patient care (10/10).
Check out the paper and appendix for many more results, including exploration of zero shot, various input parameters, latent clinical knowledge and tech details (9/10)
October 7, 2025 at 7:39 AM
Check out the paper and appendix for many more results, including exploration of zero shot, various input parameters, latent clinical knowledge and tech details (9/10)
In zero shot forecasting, DT-GPT outperformed a fully trained model on 13 variables. These variables were typically biologically linked to the target variables used during training (8/10)
October 7, 2025 at 7:38 AM
In zero shot forecasting, DT-GPT outperformed a fully trained model on 13 variables. These variables were typically biologically linked to the target variables used during training (8/10)
We show that key variables (e.g. therapy, ECOG) can drive differences in both predictions and real data. DT-GPT can even offer preliminary explainability & perform zero-shot on variables that it did not see during training. (7/10)
October 7, 2025 at 7:38 AM
We show that key variables (e.g. therapy, ECOG) can drive differences in both predictions and real data. DT-GPT can even offer preliminary explainability & perform zero-shot on variables that it did not see during training. (7/10)
DT-GPT is robust: achieves competitive performance after ~5,000 patients, can even handle a 20% increase in missingness and up to 25 misspellings per sample without significant performance degradation (6/10)
October 7, 2025 at 7:38 AM
DT-GPT is robust: achieves competitive performance after ~5,000 patients, can even handle a 20% increase in missingness and up to 25 misspellings per sample without significant performance degradation (6/10)
Taking a step back, we see that DT-GPT also preserves the overall distribution of the outputs better than other baselines, quantified with the KS-distance (5/10)
October 7, 2025 at 7:37 AM
Taking a step back, we see that DT-GPT also preserves the overall distribution of the outputs better than other baselines, quantified with the KS-distance (5/10)
Digging deeper, DT-GPT generally outperforms the second best model longitudinally. In many cases, high error predictions occur since our forecasts are based on aggregations of multiple trajectories, even if some individual trajectories are closer to the ground truth (4/10)
October 7, 2025 at 7:37 AM
Digging deeper, DT-GPT generally outperforms the second best model longitudinally. In many cases, high error predictions occur since our forecasts are based on aggregations of multiple trajectories, even if some individual trajectories are closer to the ground truth (4/10)
Our method, DT-GPT, outperforms the SOTA baselines in most cases, or achieves very competitive performance. Here you see the mean absolute error (MAE) across 12 variables in 3 different indications (3/10)
October 7, 2025 at 7:37 AM
Our method, DT-GPT, outperforms the SOTA baselines in most cases, or achieves very competitive performance. Here you see the mean absolute error (MAE) across 12 variables in 3 different indications (3/10)
We fine-tune biomedical LLMs on patient clinical data, exploring the method on both a long term lung cancer dataset, and a short term ICU dataset. A few adjustments are required to get full performance, esp. trajectory aggregation & instruction masking (2/10)
October 7, 2025 at 7:37 AM
We fine-tune biomedical LLMs on patient clinical data, exploring the method on both a long term lung cancer dataset, and a short term ICU dataset. A few adjustments are required to get full performance, esp. trajectory aggregation & instruction masking (2/10)
DT-GPT: showing that LLMs can forecast patient trajectories (1/10)
Now in npj Digital Medicine www.nature.com/articles/s41...
Also in Doctor Penguin!
Big thanks to Maria Bordukova, @raulrod.bsky.social Papichaya Quengdaeng Daniel Garger @fschmich.bsky.social Michael Menden Helmholtz Munich Roche
Now in npj Digital Medicine www.nature.com/articles/s41...
Also in Doctor Penguin!
Big thanks to Maria Bordukova, @raulrod.bsky.social Papichaya Quengdaeng Daniel Garger @fschmich.bsky.social Michael Menden Helmholtz Munich Roche
October 7, 2025 at 7:36 AM
DT-GPT: showing that LLMs can forecast patient trajectories (1/10)
Now in npj Digital Medicine www.nature.com/articles/s41...
Also in Doctor Penguin!
Big thanks to Maria Bordukova, @raulrod.bsky.social Papichaya Quengdaeng Daniel Garger @fschmich.bsky.social Michael Menden Helmholtz Munich Roche
Now in npj Digital Medicine www.nature.com/articles/s41...
Also in Doctor Penguin!
Big thanks to Maria Bordukova, @raulrod.bsky.social Papichaya Quengdaeng Daniel Garger @fschmich.bsky.social Michael Menden Helmholtz Munich Roche
Reposted by Nikita Makarov
A new model, DT-GPT, uses LLMs to forecast patient health trajectories, enabling "digital twins." By processing raw EHR data, it outperformed state-of-the-art methods in cancer, ICU, and Alzheimer's cohorts and can even forecast untrained variables.
#MedSky #MedAI #MLSky
#MedSky #MedAI #MLSky
Large language models forecast patient health trajectories enabling digital twins - npj Digital Medicine
npj Digital Medicine - Large language models forecast patient health trajectories enabling digital twins
www.nature.com
October 2, 2025 at 2:47 PM
Reposted by Nikita Makarov
[1/4] 🎉 We're thrilled to announce the general release of three de-identified, longitudinal EHR datasets from Stanford Medicine—now freely available for non-commercial research use worldwide! 🚀
Learn more on our HAI blog:
hai.stanford.edu/news/advanci...
Learn more on our HAI blog:
hai.stanford.edu/news/advanci...
Advancing Responsible Healthcare AI with Longitudinal EHR Datasets
Current evaluations of AI models in healthcare rely on limited datasets like MIMIC, lacking complete patient trajectories. New benchmark datasets offer an alternative.
hai.stanford.edu
February 13, 2025 at 1:38 AM
[1/4] 🎉 We're thrilled to announce the general release of three de-identified, longitudinal EHR datasets from Stanford Medicine—now freely available for non-commercial research use worldwide! 🚀
Learn more on our HAI blog:
hai.stanford.edu/news/advanci...
Learn more on our HAI blog:
hai.stanford.edu/news/advanci...
Want to push the limits of LLMs in drug development?
🚀Apply now for our 2 summer internships for 2025:
1️⃣ Multimodal - www.linkedin.com/jobs/view/40...
2️⃣ Preclinical - www.linkedin.com/jobs/view/40...
DM me if you have any questions or know anybody who would be interested in this.
🚀Apply now for our 2 summer internships for 2025:
1️⃣ Multimodal - www.linkedin.com/jobs/view/40...
2️⃣ Preclinical - www.linkedin.com/jobs/view/40...
DM me if you have any questions or know anybody who would be interested in this.
www.linkedin.com
December 16, 2024 at 12:19 PM
Want to push the limits of LLMs in drug development?
🚀Apply now for our 2 summer internships for 2025:
1️⃣ Multimodal - www.linkedin.com/jobs/view/40...
2️⃣ Preclinical - www.linkedin.com/jobs/view/40...
DM me if you have any questions or know anybody who would be interested in this.
🚀Apply now for our 2 summer internships for 2025:
1️⃣ Multimodal - www.linkedin.com/jobs/view/40...
2️⃣ Preclinical - www.linkedin.com/jobs/view/40...
DM me if you have any questions or know anybody who would be interested in this.
Reposted by Nikita Makarov
For my fellow researchers in AI, Medical, and Healthcare domain:, Here is the Medical AI Startup pack if you are new there
go.bsky.app/PddA2uy
go.bsky.app/PddA2uy
Medical AI
Join the conversation
go.bsky.app
November 27, 2024 at 9:13 AM
For my fellow researchers in AI, Medical, and Healthcare domain:, Here is the Medical AI Startup pack if you are new there
go.bsky.app/PddA2uy
go.bsky.app/PddA2uy
This is great! Would it be possible to add me to this? 🙏
November 27, 2024 at 9:18 AM
This is great! Would it be possible to add me to this? 🙏
Reposted by Nikita Makarov
I created a starter pack for Health AI and Informatics. Mix of folks (reporters and researchers) that I think you should follow.
I've got room to include more, so please tag anyone you think I should add! 🧪🩺 🤖 🛟
I've got room to include more, so please tag anyone you think I should add! 🧪🩺 🤖 🛟
October 25, 2024 at 3:06 PM
I created a starter pack for Health AI and Informatics. Mix of folks (reporters and researchers) that I think you should follow.
I've got room to include more, so please tag anyone you think I should add! 🧪🩺 🤖 🛟
I've got room to include more, so please tag anyone you think I should add! 🧪🩺 🤖 🛟
This is fantastic! Would it be possible to add me in as well?
November 22, 2024 at 7:05 AM
This is fantastic! Would it be possible to add me in as well?
Pre-print here: medrxiv.org/content/10.1...
“Large Language Models forecast Patient Health Trajectories enabling Digital Twins”
Reposted from X to have some content here :)
“Large Language Models forecast Patient Health Trajectories enabling Digital Twins”
Reposted from X to have some content here :)
Large Language Models forecast Patient Health Trajectories enabling Digital Twins
Background Generative artificial intelligence (AI) accelerates the development of digital twins, which enable virtual representations of real patients to explore, predict and simulate patient health t...
medrxiv.org
November 20, 2024 at 10:07 AM
Pre-print here: medrxiv.org/content/10.1...
“Large Language Models forecast Patient Health Trajectories enabling Digital Twins”
Reposted from X to have some content here :)
“Large Language Models forecast Patient Health Trajectories enabling Digital Twins”
Reposted from X to have some content here :)
Overall, DT-GPT shows that LLMs have the potential to become human digital twins. We hope that, in the future, LLM based digital twins will revolutionize the way we run clinical trials & patient care (8/8).
November 20, 2024 at 10:07 AM
Overall, DT-GPT shows that LLMs have the potential to become human digital twins. We hope that, in the future, LLM based digital twins will revolutionize the way we run clinical trials & patient care (8/8).
In zero shot forecasting, DT-GPT outperformed a fully trained model on 13 variables. These variables were typically biologically linked to the target variables used during training (7/8)
November 20, 2024 at 10:07 AM
In zero shot forecasting, DT-GPT outperformed a fully trained model on 13 variables. These variables were typically biologically linked to the target variables used during training (7/8)
DT-GPT can offer preliminary explainability & perform zero-shot on variables that it did not see during training. We show that key variables (e.g. therapy, ECOG) can drive differences in both predictions and real data (6/8)
November 20, 2024 at 10:06 AM
DT-GPT can offer preliminary explainability & perform zero-shot on variables that it did not see during training. We show that key variables (e.g. therapy, ECOG) can drive differences in both predictions and real data (6/8)