Pith. sign in

REVIEW 5 major objections 6 minor 34 references

LLM-Driven Medical Document Analysis: Enhancing Trustworthy Pathology and Differential Diagnosis

T0 review · 5 major / 6 minor · reviewed 2026-08-15 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read A LoRA-tuned LLaMA-v3, trained only on synthetic DDXPlus records, reports 99.94% GTPA on differential diagnosis and 99.81% pathology accuracy, with a local web deployment for privacy.

desk verdict A clear LoRA-LLaMA-3 study on DDXPlus that overclaims clinical utility and has unclarified evaluation details; worth refereeing with revisions. read the letter →

arxiv 2506.19702 v1 pith:L5PWJ3Z4 submitted 2025-06-24 cs.AI

Lei Kang , Xuanshuo Fu , Oriol Ramos Terrades , Javier Vazquez-Corral , Ernest Valveny , Dimosthenis Karatzas This is my paper · ORCID
classification cs.AI
keywords largelanguagemodelslow-rankadaptationdifferentialdiagnosispathologypredictionDDXPlusmedicaldocumentanalysisexplainabilityprivacy-preservingAI
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper argues that a locally deployable, parameter-efficient fine-tune of an open-weight LLM can handle two core pre-diagnosis tasks—predicting the single most likely pathology and producing a variable-length differential diagnosis list—at levels that match or beat much larger or more specialized systems. The authors fine-tune LLaMA-v3-8B-Instruct with LoRA adapters on the DDXPlus synthetic patient-record dataset and report 99.81% accuracy for pathology prediction and 99.94% GTPA for differential diagnosis, outperforming the published baselines they compare against. The practical point is privacy: because the model runs on hospital-controlled hardware rather than cloud APIs, sensitive patient text never leaves the institution, while a Flask web interface makes the tool usable by clinicians or patients. A sympathetic reader would take the paper's contribution to be a demonstration that strong diagnostic accuracy and local, explainable deployment are not mutually exclusive.

What carries the argument

The load-bearing mechanism is LoRA applied to the frozen self-attention weights of LLaMA-v3-8B-Instruct, with the update written as $W_0 + \Delta W = W_0 + BA$ where $B \in \mathbb{R}^{d \times r}$, $A \in \mathbb{R}^{r \times k}$, and rank $r=4$. LoRA keeps the backbone frozen and trains only the low-rank adapter matrices plus two linear classification heads, so task specialization costs a fraction of full fine-tuning and the resulting model can be served on a single GPU inside a hospital network. The paper also uses the DDXPlus ground-truth lists, not per-pathology probabilities, as training targets, which is why the metric GTPA—whether the true pathology appears anywhere in the predicted list—is the headline measure.

What would settle it

Take the released model and run it on a set of real electronic health records or clinician-written case notes with confirmed final diagnoses; if GTPA or pathology accuracy drops materially below the DDXPlus numbers, or if matching the truth requires per-case threshold hand-tuning like the 0.5-to-0.35 adjustment in Section 4.7, then the paper's practical-utility claim is falsified.

Watch

Extended reading notes

Core claim

The central discovery, stated on the paper's own terms, is that a low-rank adaptation of LLaMA-v3 with two task-specific classification heads—one for pathology prediction, one for variable-length differential diagnosis—achieves state-of-the-art results on the DDXPlus benchmark: 99.81% accuracy, 96.54% precision, 94.34% recall, and 94.81% F1 for pathology prediction, and 99.94% GTPA, 99.46% accuracy, 98.18% precision, 97.91% recall, and 98.01% F1 for differential diagnosis. The authors claim their method outperforms BASD, AARLC, and DDxT on differential diagnosis, and that the attention-map visualizations reveal interpretable patterns: shallow layers attend broadly, middle layers lock onto symptom keywords, and deep layers consolidate into the classification token. The failure-case analysis shows that a post-hoc threshold adjustment from 0.5 to 0.35 recovers the full ground-truth list, which the authors present as evidence that the model has learned meaningful probability estimates even though training used only the variable-length diagnosis lists, not probability labels.

Load-bearing premise

The central claim collapses if synthetic DDXPlus patient records are not a faithful proxy for real unstructured clinical documents, since all reported accuracy numbers come from that benchmark and the paper includes no validation on real patient records or clinician-curated cases.

Editorial extensions

If this is right

  • If the reported numbers hold, a single 8B model fine-tuned with rank-4 adapters can cover both single-pathology prediction and variable-length differential diagnosis at a level the authors compare favorably with prior DDXPlus systems.
  • Because the pipeline runs on local GPU hardware, deployment does not require sending patient data to external LLM APIs, which is the privacy property the paper emphasizes.
  • The attention-map analysis suggests a usable debugging path: middle-layer attention on symptom keywords separates correct from incorrect predictions in the cases shown.
  • The threshold experiment implies that the model's output probabilities are informative and could be calibrated rather than used with a fixed 0.5 cutoff; this is a direct corollary of the paper's own failure-case discussion.
  • The web platform's eight-question form is a concrete artifact that could be tested by clinicians without requiring them to write free-form prompts.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A testable extension the authors do not run: hold out real electronic health records with confirmed final diagnoses and compare model GTPA against the DDXPlus numbers; if the gap is large, the synthetic-data assumption is the culprit.
  • The threshold sensitivity in Section 4.7 suggests that in a live deployment the system would need per-symptom calibration or a decision rule that adapts threshold to predicted probability mass, something the paper leaves implicit.
  • If the synthetic-to-real transfer does hold, the same LoRA recipe may transfer to other structured clinical tasks where large labeled corpora are scarce, because DDXPlus-style synthetic generation could replace manual annotation.
  • The attention maps hint that clinicians might trust the system more by inspecting middle-layer focus on symptom phrases, but the paper does not test whether such inspection actually changes clinician decisions.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

5 major / 6 minor

Summary. The paper proposes a medical document analysis platform that fine-tunes Meta-Llama-3.1-8B-Instruct with LoRA on the DDXPlus synthetic dataset for two tasks: pathology prediction (single most likely disease) and differential diagnosis (variable-length list of possible diseases). The authors report pathology prediction accuracy of 99.81% (slightly below DDxT's 99.98%), differential diagnosis GTPA of 99.94%, precision 98.18%, recall 97.91%, and F1 98.01%, alongside a Flask-based web interface and self-attention visualizations at shallow, middle, and deep layers. The stated contributions are a locally deployable, privacy-preserving system, state-of-the-art benchmark performance, and explainability for clinical trust.

Significance. If the claims are substantiated, the work would be a useful engineering contribution: it demonstrates that a relatively small open-weight LLM can be adapted with LoRA to a large medical benchmark, that the resulting model can be served locally, and that attention maps can be inspected post hoc. Credit is due for releasing code, using a large public benchmark, and addressing privacy through local deployment. However, the practical and clinical significance is currently limited because the evaluation is confined to synthetic structured DDXPlus records, the reported metrics lack uncertainty quantification, and the stated 'state-of-the-art' claim is not supported by the pathology prediction results. The significance depends on whether the metric definitions and threshold choices are clarified and whether the method generalizes beyond the synthetic benchmark.

major comments (5)
  1. [Section 4.3, Table 3] The Accuracy formula (Eq. 1) is a binary classification formula involving TP, FP, TN, and FN, but Table 3 reports Accuracy for differential diagnosis, which produces variable-length sets of predicted diseases. The manuscript never defines how TP/FP/TN/FN are aggregated across patients or across diseases for this multi-label-style setting, so the headline value of 99.46% is not interpretable as written. Please provide the exact aggregation rule.
  2. [Section 4.4, Abstract, Conclusion] The broad claim that the proposed method 'surpasses current state-of-the-art models' is contradicted by the paper's own Table 2, where DDxT achieves 99.98% pathology prediction accuracy versus 99.81% for the proposed model. The text acknowledges this but the abstract and conclusion nevertheless assert superiority. Please restrict the superiority claim to the specific metrics where it holds, e.g., GTPA, precision, recall, and F1 for differential diagnosis, and do not claim overall pathology-prediction superiority.
  3. [Section 4.7, Figs. 6-8, Table 3] The manuscript shows that lowering the prediction threshold from 0.5 to 0.35 post hoc turns a failure case (4 of 7 ground-truth diagnoses) into a perfect match, but it never states which threshold was used to produce the results in Table 3 or whether the threshold was selected on the test set. Since the output set size and therefore all differential-diagnosis metrics depend critically on this threshold, the reported numbers may reflect test-set adaptation rather than a fixed inference rule. Please report the threshold used, justify it from validation data, and assess sensitivity of the reported metrics to threshold choice.
  4. [Sections 3.3, 4.1, 4.5; Abstract] The abstract and introduction claim the system analyzes 'unstructured medical documents' and has 'practical utility in clinical settings,' but the evaluation uses only synthetic DDXPlus records, which are structured JSON fields (sex, age, region, symptoms, antecedents), and the web interface collects exactly these structured fields (Table 1, Fig. 2). No evaluation is performed on real free-form clinical notes, electronic health records, or clinician-curated cases, so the transfer to unstructured clinical documents is unsubstantiated. Please either validate on realistic unstructured inputs or substantially temper the generalization claims.
  5. [Tables 2 and 3] The reported differences between the proposed method and the strongest baselines are very small (e.g., GTPA 99.94% vs. AARLC 99.92%), but no error bars, confidence intervals, test-set size, or significance tests are provided. The test set size is not stated anywhere in the paper, so a 0.02 percentage point GTPA difference may be within sampling noise. Please provide uncertainty estimates and, where possible, significance tests or confidence intervals that support the claimed advantage.
minor comments (6)
  1. [Section 2, Related Work] Several phrases have missing spaces, e.g., 'demonstrateditseffectivenessinfine-tuninglargelanguagemodels' and 'BuildingonthestrengthsofLoRA,MOELoRA'; the text needs a careful proofreading pass.
  2. [Section 3.1] The paper calls the model 'LLaMA-v3' but uses 'Meta-Llama-3.1-8B-Instruct'; please be precise about the exact model version, since LLaMA 3 and 3.1 have different capabilities and licenses.
  3. [Section 4.6, Figs. 4-5] The explainability analysis is qualitative and based on a single correct case and a single failure case; a quantitative evaluation of attention maps (e.g., against human annotations or as a predictor of correctness) would better support the trustworthiness claims.
  4. [Table 3] The table gives no GTPA or Accuracy for DDxT; the paper should state whether these numbers were unavailable in the original publication or simply omitted, and why the comparison is still meaningful.
  5. [Section 4.7] The text says 'we lack probability-based ground truth for training the LLM' but then interprets the model's sigmoid outputs as well-calibrated probabilities; this tension should be addressed, since without calibration the threshold discussion is heuristic.
  6. [General] The paper does not discuss de-identification, data licensing, or clinical validation ethics, despite being a medical pre-diagnosis platform; a short statement on these aspects would improve the manuscript.

Circularity Check

1 steps flagged · score 2.0 of 10

Post-hoc threshold tuning in a failure case; the central DDXPlus benchmark evaluation is independent and not circular.

  1. fitted input called prediction [Section 4.7, Discussion of the Failure Prediction Charts (Fig. 7).]
    "By simply lowering the prediction threshold from 0.5 to 0.35 as illustrated in Fig. 7, we can include 3 additional cases, perfectly matching the ground truth differential diagnosis."

    The threshold of 0.35 is selected after observing the ground-truth labels of the failure case, so the resulting 'perfect match' is guaranteed by the threshold choice rather than by the model's predictive generalization. The paper presents this as evidence that 'our proposed method can still address these failure cases,' but the match is by construction a post-hoc fit to the known ground truth. This is a minor, illustrative adjustment in the failure analysis and does not undermine the main benchmark comparison, but it should be described as threshold fitting rather than as evidence of robustness.

full rationale

The central derivation is a standard supervised learning pipeline: a LoRA-tuned LLaMA-v3 is fine-tuned on DDXPlus training records and evaluated on the DDXPlus test set. This is not circular; the model's pathology and differential-diagnosis predictions are not used as training inputs in a way that would force the reported GTPA, accuracy, precision, recall, or F1 scores. The SOTA comparisons cite published baseline numbers from external groups, and no load-bearing uniqueness or self-citation chain is used. The only identifiable circular step is in Section 4.7, where a failure-case threshold is lowered from 0.5 to 0.35 after inspecting the ground truth, making the subsequent 'perfect match' a fitted result rather than a prediction. This is a single illustrative case and does not affect the central benchmark claims, but it is a real post-hoc adjustment. Accordingly, the overall circularity score is low.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

The paper introduces no new entities. The free parameters are mostly standard hyperparameters, but the prediction threshold is a genuine free parameter because Section 4.7 shows it materially changes the outcome. The central assumption is that synthetic DDXPlus records stand in for real clinical text.

free parameters (3)
  • prediction threshold = 0.5 (lowered to 0.35 in Section 4.7)
    Used to binarize the differential diagnosis sigmoid outputs. Section 4.7 shows that lowering the threshold from 0.5 to 0.35 makes a failure case match the ground truth exactly, so the reported GTPA depends on this choice.
  • LoRA rank r = 4
    Selected without ablation or justification, typical default for LoRA fine-tuning. It conditions the expressiveness of the fine-tuned model.
  • number of training epochs = 1 for pathology, 2 for differential diagnosis
    The paper does not justify these choices; no early stopping or epoch search is reported. Different epochs could change performance.
assumptions (4)
  • domain assumption DDXPlus ground-truth labels are accurate and representative of real clinical diagnosis.
    The entire evaluation (Section 4.1) relies on the synthetic patient records and their assigned diagnoses. No real clinical corpus is used for validation.
  • domain assumption Attention maps are a valid and sufficient explanation of model predictions.
    Section 4.6 treats attention visualizations as evidence of trustworthy decision-making, but attention maps are not yet established as faithful explanations in this context, and only two examples are shown.
  • standard math The binary Accuracy formula (TP+TN)/(TP+FP+TN+FN) is appropriate for a multi-class pathology prediction task.
    Section 4.3 defines Accuracy using binary classification terms despite pathology prediction being multi-class (49 classes). The paper does not specify if this is macro-averaged or global, making the reported 99.81% ambiguous.
  • standard math LoRA weight updates maintain the intrinsic low-rank hypothesis of Aghajanyan et al. and Hu et al.
    Section 3.2 cites these works as the theoretical basis for the fine-tuning method; the paper does not verify the low-rank assumption for medical data.

how reviews work

0 comments
Cite this review

Pith. "Pith review of LLM-Driven Medical Document Analysis: Enhancing Trustworthy Pathology and Differential Diagnosis." pith.science (2026). https://pith.science/paper/L5PWJ3Z4

@misc{pith2026250619702,
  author       = {Pith},
  title        = {Pith review of: LLM-Driven Medical Document Analysis: Enhancing Trustworthy Pathology and Differential Diagnosis},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/L5PWJ3Z4}},
  note         = {Machine review of arXiv:2506.19702}
}
read the original abstract

Medical document analysis plays a crucial role in extracting essential clinical insights from unstructured healthcare records, supporting critical tasks such as differential diagnosis. Determining the most probable condition among overlapping symptoms requires precise evaluation and deep medical expertise. While recent advancements in large language models (LLMs) have significantly enhanced performance in medical document analysis, privacy concerns related to sensitive patient data limit the use of online LLMs services in clinical settings. To address these challenges, we propose a trustworthy medical document analysis platform that fine-tunes a LLaMA-v3 using low-rank adaptation, specifically optimized for differential diagnosis tasks. Our approach utilizes DDXPlus, the largest benchmark dataset for differential diagnosis, and demonstrates superior performance in pathology prediction and variable-length differential diagnosis compared to existing methods. The developed web-based platform allows users to submit their own unstructured medical documents and receive accurate, explainable diagnostic results. By incorporating advanced explainability techniques, the system ensures transparent and reliable predictions, fostering user trust and confidence. Extensive evaluations confirm that the proposed method surpasses current state-of-the-art models in predictive accuracy while offering practical utility in clinical settings. This work addresses the urgent need for reliable, explainable, and privacy-preserving artificial intelligence solutions, representing a significant advancement in intelligent medical document analysis for real-world healthcare applications. The code can be found at \href{https://github.com/leitro/Differential-Diagnosis-LoRA}{https://github.com/leitro/Differential-Diagnosis-LoRA}.

Figures

Figures reproduced from arXiv: 2506.19702 by the authors.

Figure 1
Figure 1. The proposed medical document analysis architecture for pathology and [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. The proposed medical pre-diagnosis platform pipeline. [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗
Figure 3
Figure 3. Visualization of the website platform. 4.3 Metrics We evaluate the effectiveness of our method using several widely recognized performance metrics: Accuracy, Precision, Recall, F1 score, and Ground Truth Pathology Accuracy (GTPA) [7]. Accuracy is defined as: Accuracy = T P + T N T P + F P + T N + F N where T P (True Positives), F P (False Positives), T N (True Negatives), and F N (False Negatives) are the counts of … view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: Self-Attention visual￾ization for Shallow, Middle and Deep layers in LoRA￾tuned LLaMA-v3 model for correct pathology prediction for “Cluster headache”. (a) Shallow Layer (b) Middle Layer (c) Deep Layer [PITH_FULL_IMAGE:figures/full_fig_p012_4.png]
Figure 6
Figure 6. Figure 6: A failure case where the differential diagnosis predicts only 4 cases, while [PITH_FULL_IMAGE:figures/full_fig_p013_6.png]
Figure 7
Figure 7. Figure 7: By lowering the prediction threshold from 0.5, three additional cases are [PITH_FULL_IMAGE:figures/full_fig_p013_7.png]
Figure 8
Figure 8. Figure 8: Displaying all 49 disease categories for differential diagnosis. [PITH_FULL_IMAGE:figures/full_fig_p013_8.png]

Review history (2 revisions) →

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

34 extracted references · 17 canonical work pages

  1. [1]

    arXiv preprint arXiv:2303.08774 (2023)

    Achiam, J., Adler, S., Agarwal, S., Ahmad, L., Akkaya, I., Aleman, F.L., Almeida, D., Altenschmidt, J., Altman, S., Anadkat, S., et al.: Gpt-4 technical report. arXiv preprint arXiv:2303.08774 (2023)

    arXiv 2023

  2. [2]

    Aghajanyan, A., Zettlemoyer, L., Gupta, S.: Intrinsic dimensionality explains the effectivenessoflanguagemodelfine-tuning.arXivpreprintarXiv:2012.13255(2020)

    arXiv 2020

  3. [3]

    In: Deep Generative Models for Health Workshop NeurIPS 2023 (2023)

    Alam, M.M., Raff, E., Oates, T., Matuszek, C.: Ddxt: Deep generative transformer models for differential diagnosis. In: Deep Generative Models for Health Workshop NeurIPS 2023 (2023)

    work page 2023

  4. [4]

    Journal of Machine Learning Research24(240), 1–113 (2023)

    Chowdhery, A., Narang, S., Devlin, J., Bosma, M., Mishra, G., Roberts, A., Barham, P., Chung, H.W., Sutton, C., Gehrmann, S., et al.: Palm: Scaling lan- guage modeling with pathways. Journal of Machine Learning Research24(240), 1–113 (2023)

    2023

  5. [5]

    Journal of Machine Learning Research25(70), 1–53 (2024)

    Chung, H.W., Hou, L., Longpre, S., Zoph, B., Tay, Y., Fedus, W., Li, Y., Wang, X., Dehghani, M., Brahma, S., et al.: Scaling instruction-finetuned language models. Journal of Machine Learning Research25(70), 1–53 (2024)

    2024

  6. [6]

    arXiv preprint arXiv:2407.21783 (2024)

    Dubey, A., Jauhri, A., Pandey, A., Kadian, A., Al-Dahle, A., Letman, A., Mathur, A., Schelten, A., Yang, A., Fan, A., et al.: The llama 3 herd of models. arXiv preprint arXiv:2407.21783 (2024)

    arXiv 2024

  7. [7]

    Advances in neural information processing systems 35, 31306–31318 (2022)

    Fansi Tchango, A., Goel, R., Wen, Z., Martel, J., Ghosn, J.: Ddxplus: A new dataset for automatic medical diagnosis. Advances in neural information processing systems 35, 31306–31318 (2022)

    work page 2022

  8. [8]

    Annals of Biomedical Engineering52(5), 1119– 1124 (2024)

    Ferdush, J., Begum, M., Hossain, S.T.: Chatgpt and clinical decision support: scope, application, and limitations. Annals of Biomedical Engineering52(5), 1119– 1124 (2024)

    work page 2024

Show all 34 references
  1. [9]

    Health Promotion Perspectives13(3), 183 (2023)

    Garg, R.K., Urs, V.L., Agarwal, A.A., Chaudhary, S.K., Paliwal, V., Kar, S.K.: Exploring the role of chatgpt in patient care (diagnosis and treatment) and medical research: A systematic review. Health Promotion Perspectives13(3), 183 (2023)

    2023

  2. [10]

    Computers in biology and medicine 155, 106649 (2023)

    Hossain, E., Rana, R., Higgins, N., Soar, J., Barua, P.D., Pisani, A.R., Turner, K.: Natural language processing in electronic health records in relation to healthcare decision-making: a systematic review. Computers in biology and medicine 155, 106649 (2023)

    2023

  3. [11]

    arXiv preprint arXiv:2106.09685 (2021)

    Hu, E.J., Shen, Y., Wallis, P., Allen-Zhu, Z., Li, Y., Wang, S., Wang, L., Chen, W.: Lora: Low-rank adaptation of large language models. arXiv preprint arXiv:2106.09685 (2021)

    2021 arXiv

  4. [12]

    In: Proceedings of the IEEE/CVF international conference on computer vision

    Huang, S.C., Shen, L., Lungren, M.P., Yeung, S.: Gloria: A multimodal global-local representation learning framework for label-efficient medical image recognition. In: Proceedings of the IEEE/CVF international conference on computer vision. pp. 3942–3951 (2021) 16 L. Kang et al

    2021

  5. [13]

    Applied Sciences11(14), 6421 (2021)

    Jin, D., Pan, E., Oufattole, N., Weng, W.H., Fang, H., Szolovits, P.: What disease does this patient have? a large-scale open domain question answering dataset from medical exams. Applied Sciences11(14), 6421 (2021)

    2021

  6. [14]

    arXiv preprint arXiv:1909.06146 (2019)

    Jin, Q., Dhingra, B., Liu, Z., Cohen, W.W., Lu, X.: Pubmedqa: A dataset for biomedical research question answering. arXiv preprint arXiv:1909.06146 (2019)

    2019 arXiv

  7. [15]

    IEEE transactions on visualization and computer graphics 25(1), 299–309 (2018)

    Kwon, B.C., Choi, M.J., Kim, J.T., Choi, E., Kim, Y.B., Kwon, S., Sun, J., Choo, J.: Retainvis: Visual analytics with interpretable and interactive recurrent neural networks on electronic medical records. IEEE transactions on visualization and computer graphics 25(1), 299–309 (2018)

    2018

  8. [16]

    arXiv preprint arXiv:2407.19299 (2024)

    Le, T.D., Nguyen, T.T., Ha, V.N.: The impact of lora adapters for llms on clinical nlp classification under data limitations. arXiv preprint arXiv:2407.19299 (2024)

    2024 arXiv

  9. [17]

    Bioinformatics 36(4), 1234–1240 (2020)

    Lee, J., Yoon, W., Kim, S., Kim, D., Kim, S., So, C.H., Kang, J.: Biobert: a pre-trained biomedical language representation model for biomedical text mining. Bioinformatics 36(4), 1234–1240 (2020)

    2020

  10. [18]

    In: Proceedings of the 47th International ACM SIGIR Conference on Research and Development in Information Retrieval

    Liu, Q., Wu, X., Zhao, X., Zhu, Y., Xu, D., Tian, F., Zheng, Y.: When moe meets llms: Parameter efficient fine-tuning for multi-task medical applications. In: Proceedings of the 47th International ACM SIGIR Conference on Research and Development in Information Retrieval. pp. 1...

    2024

  11. [19]

    arXiv preprint arXiv:2101.09773 (2021)

    Luo, H., Li, S.W., Glass, J.: Knowledge grounded conversational symptom detec- tion with graph memory networks. arXiv preprint arXiv:2101.09773 (2021)

    2021 arXiv

  12. [20]

    Briefings in bioinformatics 23(6), bbac409 (2022)

    Luo,R.,Sun,L.,Xia,Y.,Qin,T.,Zhang,S.,Poon,H.,Liu,T.Y.:Biogpt:generative pre-trained transformer for biomedical text generation and mining. Briefings in bioinformatics 23(6), bbac409 (2022)

    2022

  13. [21]

    In: 2024 IEEE 12th International Conference on Healthcare Informatics (ICHI)

    McPeak, G., Sautmann, A., George, O., Hallal, A., Simal, E.A., Schwartz, A.L., Abaluck, J., Ravi, N., Pless, R.: An llm’s medical testing recommendations in a nigerian clinic: Potential and limits of prompt engineering for clinical decision support. In: 2024 IEEE 12th Internat...

    2024

  14. [22]

    In: Conference on health, inference, and learning

    Pal, A., Umapathi, L.K., Sankarasubbu, M.: Medmcqa: A large-scale multi-subject multi-choice dataset for medical domain question answering. In: Conference on health, inference, and learning. pp. 248–260. PMLR (2022)

    2022

  15. [23]

    Journal of Machine Learning for Healthcare Decision Support 3(1), 18–62 (2023)

    Prabhod, K.J.: Integrating large language models for enhanced clinical decision support systems in modern healthcare. Journal of Machine Learning for Healthcare Decision Support 3(1), 18–62 (2023)

    2023

  16. [24]

    arXiv preprint arXiv:2402.01711 (2024)

    Schmiedmayer, P., Rao, A., Zagar, P., Ravi, V., Zahedivash, A., Fereydooni, A., Aalami, O.: Llm on fhir–demystifying health records. arXiv preprint arXiv:2402.01711 (2024)

    2024 arXiv

  17. [25]

    Nature 620(7972), 172–180 (2023)

    Singhal, K., Azizi, S., Tu, T., Mahdavi, S.S., Wei, J., Chung, H.W., Scales, N., Tanwani, A., Cole-Lewis, H., Pfohl, S., et al.: Large language models encode clinical knowledge. Nature 620(7972), 172–180 (2023)

    2023

  18. [26]

    arXiv preprint arXiv:2305.09617 (2023)

    Singhal, K., Tu, T., Gottweis, J., Sayres, R., Wulczyn, E., Hou, L., Clark, K., Pfohl, S., Cole-Lewis, H., Neal, D., et al.: Towards expert-level medical question answering with large language models. arXiv preprint arXiv:2305.09617 (2023)

    2023 arXiv

  19. [27]

    arXiv preprint arXiv:2302.13971 (2023)

    Touvron, H., Lavril, T., Izacard, G., Martinet, X., Lachaux, M.A., Lacroix, T., Rozière, B., Goyal, N., Hambro, E., Azhar, F., et al.: Llama: Open and efficient foundation language models. arXiv preprint arXiv:2302.13971 (2023)

    2023 arXiv

  20. [28]

    Diagnostic pathology 19(1), 43 (2024) LLM-Driven Medical Document Analysis 17

    Ullah, E., Parwani, A., Baig, M.M., Singh, R.: Challenges and barriers of using large language models (llm) such as chatgpt for diagnostic medicine with a focus on digital pathology–a recent scoping review. Diagnostic pathology 19(1), 43 (2024) LLM-Driven Medical Document Analysis 17

    2024

  21. [29]

    arXiv preprint arXiv:2401.16107 (2024)

    Wang, H., Zhao, S., Qiang, Z., Xi, N., Qin, B., Liu, T.: Beyond direct diagno- sis: Llm-based multi-specialist agent consultation for automatic diagnosis. arXiv preprint arXiv:2401.16107 (2024)

    2024 arXiv

  22. [30]

    Communications Engineering 3(1), 133 (2024)

    Wang, S., Zhao, Z., Ouyang, X., Liu, T., Wang, Q., Shen, D.: Interactive computer- aided diagnosis on medical image using large language models. Communications Engineering 3(1), 133 (2024)

    2024

  23. [31]

    Journal of the American Med- ical Informatics Association p

    Wu, C., Lin, W., Zhang, X., Zhang, Y., Xie, W., Wang, Y.: Pmc-llama: toward building open-source language models for medicine. Journal of the American Med- ical Informatics Association p. ocae045 (2024)

    2024

  24. [32]

    arXiv preprint arXiv:2402.12749 (2024)

    Xie, Q., Chen, Q., Chen, A., Peng, C., Hu, Y., Lin, F., Peng, X., Huang, J., Zhang, J., Keloth, V., et al.: Me llama: Foundation large language models for medical applications. arXiv preprint arXiv:2402.12749 (2024)

    2024 arXiv

  25. [33]

    Artificial Intelligence in Medicine148, 102748 (2024)

    Yuan, H., Yu, S.: Efficient symptom inquiring and diagnosis via adaptive alignment of reinforcement learning and classification. Artificial Intelligence in Medicine148, 102748 (2024)

    2024

  26. [34]

    In: American Medical Informatics Association (AMIA) Annual Symposium (2023)

    Yuan, J., Tang, R., Jiang, X., Hu, X.: Llm for patient-trial matching: Privacy-aware data augmentation towards better performance and generalizability. In: American Medical Informatics Association (AMIA) Annual Symposium (2023)

    2023

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.