Scaling Arabic Medical Chatbots Using Synthetic Data: Enhancing Generative AI with Synthetic Patient Records

Abdulrahman Allam; Ali Hamdi; Khaled Shaban; Seif Ahmed

REVIEW 3 major objections 5 minor 23 references

Scaling Arabic Medical Chatbots Using Synthetic Data: Enhancing Generative AI with Synthetic Patient Records

T0 review · 3 major / 5 minor · reviewed 2026-08-04 · deepseek-v4-flash

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

Pith's one-line read Synthetic patient–doctor pairs, 80,000 of them, pushed the best Arabic medical chatbot's BERTScore F1 from 68.5% to 81.4%.

desk verdict Useful Arabic-medical synthetic-data recipe undermined by an undefined evaluation split; the central comparison needs a real-only held-out test before the numbers can be trusted. read the letter →

arxiv 2509.10108 v1 pith:LFRDSYMP submitted 2025-09-12 cs.CL

Abdulrahman Allam , Seif Ahmed , Ali Hamdi , Khaled Shaban This is my paper

classification cs.CL

keywords syntheticdataArabicNLPmedicalchatbotfine-tuningBERTScorelargelanguagemodelslow-resourcelanguagesaugmentation

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 tries to establish that large-scale synthetic data can substitute for scarce real-world Arabic medical conversations. The authors expanded a 20,000-record dataset of real patient–doctor exchanges to 100,000 records by generating 80,000 synthetic question-answer pairs with ChatGPT-4o and Gemini 2.5 Pro, then fine-tuned five language models. On the BERTScore F1 metric, every model improved, with Mistral-7B jumping from 68.50% to 81.36%. An ablation study shows ChatGPT-4o-generated data outperforms Gemini-generated data for every model, pointing to data source quality as a key lever. If true, synthetic augmentation offers a practical path to more capable Arabic medical chatbots without collecting more private patient data.

What carries the argument

The key mechanism is a synthetic data generation pipeline: seed prompts anchored to real interactions, generation by two large generative models (ChatGPT-4o and Gemini 2.5 Pro), cosine-similarity and embedding-based semantic filtering, language verification, and manual review of a 500-sample subset. The augmented 100K corpus is then used to fine-tune models with low-rank adaptation (LoRA), which keeps training feasible on modest hardware. The pipeline's distinct stages—contextual anchoring, filtering, and source-specific training—are what carry the argument that synthetic data quality, not just quantity, drives downstream gains.

What would settle it

Train a model on the same 80,000 synthetic samples plus 20K real, evaluate on a held-out set of 5,000 real patient–doctor dialogues not used in training or validation, and compare BERTScore F1 against the 20K-real-only baseline. If the synthetic-augmented model fails to beat the baseline on that fully real held-out set, the claim that synthetic data improves generalization is falsified.

Watch

Extended reading notes

Core claim

The central claim is that synthetic patient–doctor dialogues, generated from seed examples and filtered for semantic relevance, can effectively augment a small real corpus enough to measurably improve model output quality. The paper demonstrates this by fine-tuning five LLMs on the 100K corpus and reporting higher BERTScore F1 than the 20K baseline for all models. Beyond raw scores, qualitative review indicates improved fluency, reduced overfitting, and fewer hallucinations, with the ChatGPT-4o source yielding the largest gains and fewer hallucinations than Gemini in the ablation.

Load-bearing premise

The improvements are measured on a validation split taken from the combined real-plus-synthetic dataset, so the reported gains may come from the model matching synthetic patterns it already saw, rather than from genuinely better medical reasoning on unseen real inputs.

Editorial extensions

If this is right

Synthetic augmentation raised BERTScore F1 for all five tested models, with Mistral-7B improving by about 13 percentage points over the real-only 20K model.
ChatGPT-4o-generated data consistently outperformed Gemini 2.5 Pro data in the same-size ablation, implying generator choice matters at least as much as volume.
Smaller models like AraGPT2-Base also gained substantially, suggesting synthetic data can help in compute-limited deployments.
The pipeline reduces overfitting and response repetition relative to 20K-only training, per qualitative review.
Validation-based early stopping and checkpoints selected by F1 improvement indicate the augmented corpus leads to better validation-set performance under the same hyperparameters.

Reading between the lines

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

If the validation split is drawn from the same combined real-plus-synthetic corpus, part of the reported gain may reflect distribution overlap rather than true generalization; a fully held-out real-only test set would settle this.
The same pipeline could be tested for other low-resource languages and clinical subdomains, where synthetic anchoring to a small real seed set may similarly amplify fine-tuning.
Because ChatGPT-4o data dominated in the ablation, prompt engineering and sampling strategy likely matter more than model scale; tracking prompt diversity and filtering thresholds as independent variables could yield a more controllable augmentation recipe.
Medical safety remains unaddressed: BERTScore and human review measure fluency and plausibility, not clinical correctness; an expert-verified safety benchmark would be needed before deployment.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Desk Editor's Note

read the letter

First, what's worth knowing: this paper builds a 100K Arabic patient-doctor corpus by generating 80K synthetic QA pairs with ChatGPT-4o and Gemini 2.5 Pro, fine-tunes five models with LoRA, and reports BERTScore F1 gains across the board, with Mistral going from ~68.5 to 81.4. The ablation between the two synthetic sources is the most distinctive part: ChatGPT-4o data consistently beats Gemini data for every model.

Credit where due: it's a concrete application to a genuinely low-resource setting, with careful preprocessing (normalization, dedup, semantic filtering) and a manual review of 500 samples. The authors are open about the challenges and cite relevant literature. The pipeline is described well enough to reproduce if the data and prompts were released.

But the evaluation has a load-bearing flaw. Section V.B says models were evaluated on a validation split from the combined dataset. If that split is a random sample of the 100K corpus, about 80% of the evaluation instances are synthetic outputs from the same two generative models used for training. BERTScore then measures how well the model reproduces the synthetic distribution, not how it performs on unseen real patient questions. The 5x training-set size alone would push scores up on a random split of the same corpus. A real-only held-out test set is necessary, and it's not there. Also, Section V.D claims three seeds and standard deviations, but no variance estimates appear in Tables I or II. The hallucination-reduction claim is qualitative only. No data, prompts, or code are released.

These are addressable omissions, but they mean the headline result—that synthetic augmentation improves Arabic medical chatbots—is not established by this paper. The ablation comparison is similarly confounded: training on ChatGPT-4o data and evaluating on a split that contains ChatGPT-4o samples will favor ChatGPT-4o.

If the authors produce a real-only held-out test, report variances, and release artifacts, the result could be worth citing. As it stands, I'd describe it as a promising engineering write-up with an evaluation that needs to be redone. Who should read it: people working on synthetic data for low-resource medical NLP, and anyone who wants a concrete example of how easy it is to leak synthetic distribution into evaluation. I'd send it to review—the question is real and the flaws are fixable—but I'd expect major revision.

Referee Report

3 major / 5 minor

Summary. The paper addresses Arabic medical chatbot development under data scarcity. The authors start from a curated 20,000-record real patient–doctor interaction dataset and propose scaling it to 100,000 records by adding 80,000 synthetic question–answer pairs generated with ChatGPT-4o and Gemini 2.5 Pro. They fine-tune five LLMs (Mistral-7B-Instruct-v0.2, LLaMA-2-7B, AraGPT2-Base, GPT-2 Medium, BLOOM-560M) with LoRA, comparing BERTScore F1 on the 20K real-only baseline against the 100K real-plus-synthetic setup, and also run an ablation comparing the two synthetic data sources. The headline result is that all five models improve with synthetic augmentation, with Mistral-7B reaching 81.36% F1 (up from 68.50% on 20K real), and that ChatGPT-4o data outperform Gemini 2.5 Pro data. The authors additionally report qualitative expert assessments indicating improved fluency, reduced hallucination, and stronger medical reasoning.

Significance. If the reported gains were backed by a valid evaluation protocol, this would be a useful contribution to low-resource Arabic medical NLP: it demonstrates a concrete pipeline for synthetic augmentation of a real seed corpus, including prompt engineering, semantic filtering, language verification, and manual review. The study addresses an under-explored area and includes a careful ablation isolating the two synthetic generation sources. However, the manuscript currently does not define the evaluation set in a way that supports the generalization claim, and it reports no variance estimates despite claiming that standard deviations were computed. These issues are fixable but are load-bearing for the central claim that synthetic data improves generalization rather than merely reproducing the synthetic training distribution.

major comments (3)

[§V.B (Training Configuration) and §VI (Results)] The evaluation set is never defined. Section V.B states that models were evaluated 'using a validation split from the combined dataset,' which after augmentation is 80% synthetic. If this split is a random sample of the 100K corpus, about 80% of the evaluation instances are synthetic outputs from ChatGPT-4o and Gemini 2.5 Pro, generated with contextual anchoring on the same real seed data used to build the training corpus. In that case, the F1 improvements in Tables I and II may reflect distribution overlap with the synthetic training set rather than generalization to unseen real patient questions. The paper must either evaluate on a held-out real-only set that was never used for generation or fine-tuning, or explicitly justify why a mixed split is not circular. Without this, the central claim is not established.
[§V.D and Tables I–II] Section V.D claims that 'all experiments were repeated three times using different random seeds' and that 'standard deviation was reported.' However, Tables I and II contain only single point estimates, with no standard deviations, confidence intervals, or significance tests. The text in §VII.B calls the ablation trend 'statistically consistent,' but this is unverifiable from the reported results. Please report the variance across seeds (e.g., mean ± std for each configuration) or remove the claim that standard deviations were reported.
[§VI.B, §VII.B, and Abstract] The qualitative claims—'improved linguistic fluency,' 'reduced overfitting,' 'stronger medical reasoning,' and especially 'fewer hallucinations'—are presented as findings but are not measured or described with a reproducible protocol. The abstract repeats the hallucination-reduction claim. Specify the number of annotators, the annotation instrument, inter-annotator agreement, and any quantification (e.g., hallucination counts per model). As written, these conclusions cannot be checked.

minor comments (5)

[§IV.A.1] Typos and formatting: 'ChatGPT-4oby OpenAI andGemini 2.5 Proby Google' is missing spaces; similar issues occur elsewhere (e.g., 'PEFT' spacing, 'Figure 2' caption).
[§IV.A.2] The semantic filtering step mentions cosine similarity filters and BERT-based embedding analysis but does not specify the threshold or the embedding model used. Please provide these details for reproducibility.
[§V.B vs. §IV.C.1] Batch size is reported as 'Tuned between 8 and 16' in §IV.C.1 and as 'Set to 8' in §V.B. Clarify which setting was actually used in the experiments.
[Tables I and II] The 'Base Model' rows are not clearly defined. State whether these are zero-shot/one-shot generations, and if so, what prompting or decoding configuration was used.
[References] Reference formatting is inconsistent (e.g., [19]–[23] are arXiv preprints with varying styles). Please align with the journal's citation style.

Circularity Check

2 steps flagged · score 6.0 of 10

Validation split from the combined real+synthetic corpus makes the reported BERTScore gains a measure of training-distribution overlap rather than independent generalization.

fitted input called prediction [Section IV.C.1 'Training Configuration'; Section V.B 'Training Configuration']
"Models were evaluated at the end of each epoch using a validation split from the combined dataset."

The headline comparison is between models fine-tuned on 20K real records and models fine-tuned on the 100K combined real+synthetic corpus. The evaluation is performed on 'a validation split from the combined dataset,' i.e., a split of the same 100K corpus that includes 80K synthetic QA pairs generated by ChatGPT-4o and Gemini 2.5 Pro. For an augmented model, this validation distribution is the same distribution used to construct its training set, so the BERTScore F1 gain largely reflects reproduction of the synthetic training distribution rather than performance on unseen real patient questions. A model trained on 100K samples would be expected to score higher on a random split of that same corpus even if the synthetic data contributed no real medical value, making the central generalizati
other [Section IV.A.2 'Data Generation Process']
"Contextual Anchoring: Each prompt was in context linked using seed examples from the original dataset to ensure coherence in medical logic and consistency with regional Arabic phrasing."

The synthetic samples are generated by anchoring prompts to seed examples from the original 20,000 real interactions. Since the validation split is taken from the combined dataset (per Section IV.C.1), any synthetic validation instances are derived from real seed examples that are themselves part of the augmented training set. This creates a leakage path independent of the train/validation split: the augmented model can score higher on validation by matching patterns it memorized from seed-derived synthetic data, so the F1 improvement does not establish generalization to genuinely new patient inputs.

full rationale

There is no equation-level derivation or fitted-parameter circularity here; the paper is an empirical NLP study. However, the central claim—that adding 80K synthetic QA pairs to 20K real Arabic interactions improves BERTScore F1—rests on an evaluation set that is not independent of the training distribution. The paper explicitly says validation is done on 'a validation split from the combined dataset,' meaning the validation set includes synthetic samples generated by the same models and from the same seed-distribution used to produce the augmented training data. Consequently, the reported gains (e.g., Mistral-7B from 68.50% to 81.36%) are partly forced by distribution overlap: an augmented model is being evaluated on the same synthetic style it was trained on, rather than on held-out real patient queries. The 'Contextual Anchoring' step makes this worse by deriving synthetic prompts from seed examples in the original 20K, which are themselves in the augmented training set. The paper does not report a real-only held-out test or any external benchmark; without that, the generalization claim is not established. Self-citations to the authors' prior work are not load-bearing for this circularity—the problem is the evaluation design, not the literature grounding. A score of 6 reflects partial circularity: the central measured improvement reduces, to a substantial degree, to train/eval distribution overlap rather than to independent evidence of medical reasoning.

Assumptions & free parameters 6 free parameters · 4 assumptions · 0 invented entities

No new physical or conceptual entities are introduced. The main assumptions are about data validity, the usefulness of synthetic medical text, and the validity of BERTScore as an evaluation proxy in this setting.

free parameters (6)

synthetic_to_real_ratio = 80k synthetic : 20k real
Chosen by hand; the 4:1 ratio directly defines the 100k corpus and is central to all reported comparisons.
semantic_filtering_threshold
Cosine similarity and BERT-based embedding filters are mentioned but no threshold is reported, so the filtering step is not reproducible.
learning_rate = 5e-5
Training hyperparameter chosen by hand, with cosine decay and 200-step warmup; affects all fine-tuned models.
epochs = 3 to 5
Chosen by model size and convergence; no per-model table is given.
batch_size = 8
Section V.B states batch size 8, while Section IV.C says 8 to 16; the discrepancy makes exact reproduction ambiguous.
loRA_rank_and_alpha
LoRA is used but rank and alpha are not reported; these settings directly affect fine-tuning behavior and reproducibility.

assumptions (4)

domain assumption The 20,000 social-media patient-doctor interactions are sufficiently accurate ground truth for medical QA.
The entire pipeline uses these interactions as seed truth, but their medical reliability is not validated against any clinical gold standard.
domain assumption Synthetic QA pairs from ChatGPT-4o and Gemini are medically coherent and safe enough for training after filtering.
This is the central premise of the paper; only 500 of 80,000 samples were manually reviewed, and no systematic clinical safety check is described.
domain assumption BERTScore with unnamed contextual embeddings is a valid semantic quality metric for Arabic medical text.
The paper does not specify which Arabic BERT embedding model was used or provide evidence that BERTScore correlates with medical correctness.
domain assumption Manual review by native Arabic speakers with basic medical knowledge is sufficient to filter harmful or misleading content.
Reviewer qualifications are described only briefly (Section IV.A.2), with no inter-annotator agreement or structured medical checklist.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Scaling Arabic Medical Chatbots Using Synthetic Data: Enhancing Generative AI with Synthetic Patient Records." pith.science (2026). https://pith.science/paper/LFRDSYMP

@misc{pith2026250910108,
  author       = {Pith},
  title        = {Pith review of: Scaling Arabic Medical Chatbots Using Synthetic Data: Enhancing Generative AI with Synthetic Patient Records},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/LFRDSYMP}},
  note         = {Machine review of arXiv:2509.10108}
}

read the original abstract

The development of medical chatbots in Arabic is significantly constrained by the scarcity of large-scale, high-quality annotated datasets. While prior efforts compiled a dataset of 20,000 Arabic patient-doctor interactions from social media to fine-tune large language models (LLMs), model scalability and generalization remained limited. In this study, we propose a scalable synthetic data augmentation strategy to expand the training corpus to 100,000 records. Using advanced generative AI systems ChatGPT-4o and Gemini 2.5 Pro we generated 80,000 contextually relevant and medically coherent synthetic question-answer pairs grounded in the structure of the original dataset. These synthetic samples were semantically filtered, manually validated, and integrated into the training pipeline. We fine-tuned five LLMs, including Mistral-7B and AraGPT2, and evaluated their performance using BERTScore metrics and expert-driven qualitative assessments. To further analyze the effectiveness of synthetic sources, we conducted an ablation study comparing ChatGPT-4o and Gemini-generated data independently. The results showed that ChatGPT-4o data consistently led to higher F1-scores and fewer hallucinations across all models. Overall, our findings demonstrate the viability of synthetic augmentation as a practical solution for enhancing domain-specific language models in-low resource medical NLP, paving the way for more inclusive, scalable, and accurate Arabic healthcare chatbot systems.

Figures

Figures reproduced from arXiv: 2509.10108 by the authors.

**Figure 2.** A sample from ChatGPT 4o Synthetic Data. [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

23 extracted references · 4 linked inside Pith

[1]

D. E. Detmer, E. B. Steen, and R. S. Dick,The computer-based patient record: an essential technology for health care. National Academy Press, 1997

1997
[2]

The concept of value in resource allocation,

T. C. Brown, “The concept of value in resource allocation,”Land economics, vol. 60, no. 3, pp. 231–246, 1984

1984
[3]

Ai-assisted diagnosis and treatment plan- ning—a discussion of how ai can assist healthcare professionals in making more accurate diagnoses and treatment plans for diseases,

M. Rashid and M. Sharma, “Ai-assisted diagnosis and treatment plan- ning—a discussion of how ai can assist healthcare professionals in making more accurate diagnoses and treatment plans for diseases,” inAI in Disease Detection: Advancements and Applications, 2025, pp. 313– 336

2025
[4]

Artificial intel- ligence and machine learning,

N. K ¨uhl, M. Schemmer, M. Goutier, and G. Satzger, “Artificial intel- ligence and machine learning,”Electronic Markets, vol. 32, no. 4, pp. 2235–2244, 2022

2022
[5]

Hospital management system with chatbot,

M. L. PRIY ANKA, B. T. R. NAY AKA, B. S. RAKSHITHA, B. V AMSHI, and B. M. K. Y ADA V , “Hospital management system with chatbot,”International Journal of Mechanical Engineering Research and Technology, vol. 16, no. 2, pp. 144–154, 2024

2024
[6]

Large language models in healthcare and medical domain: A review,

Z. A. Nazi and W. Peng, “Large language models in healthcare and medical domain: A review,”Informatics, vol. 11, no. 3, p. 57, 2024

2024
[7]

A study of generative large language model for medical research and healthcare,

C. Peng, X. Yang, A. Chen, K. E. Smith, N. PourNejatian, A. B. Costa, C. Martin, M. G. Flores, Y . Zhang, T. Magocet al., “A study of generative large language model for medical research and healthcare,” NPJ digital medicine, vol. 6, no. 1, p. 210, 2023

2023
[8]

Deep learning for arabic healthcare: Medicalbot,

M. Abdelhay, A. Mohammed, and H. A. Hefny, “Deep learning for arabic healthcare: Medicalbot,”Social Network Analysis and Mining, vol. 13, no. 1, p. 71, 2023

2023

Show all 23 references

[9]

Nabiha: an arabic dialect chat- bot,

D. Al-Ghadhban and N. Al-Twairesh, “Nabiha: an arabic dialect chat- bot,”International Journal of Advanced Computer Science and Appli- cations, vol. 11, no. 3, 2020

2020
[10]

Survey on synthetic data generation, evaluation methods and gans,

A. Figueira and B. Vaz, “Survey on synthetic data generation, evaluation methods and gans,”Mathematics, vol. 10, no. 15, p. 2733, 2022

2022
[11]

Machine learning for synthetic data generation: a review,

Y . Lu, M. Shen, H. Wang, X. Wang, C. van Rechem, T. Fu, and W. Wei, “Machine learning for synthetic data generation: a review,” arXiv preprint arXiv:2302.04062, 2023

2023 arXiv
[12]

Synthetic data generation: State of the art in health care domain,

H. Murtaza, M. Ahmed, N. F. Khan, G. Murtaza, S. Zafar, and A. Bano, “Synthetic data generation: State of the art in health care domain,” Computer Science Review, vol. 48, p. 100546, 2023

2023
[13]

Synsys: A synthetic data generation system for healthcare applications,

J. Dahmen and D. Cook, “Synsys: A synthetic data generation system for healthcare applications,”Sensors, vol. 19, no. 5, p. 1181, 2019

2019
[14]

Fake it till you make it: Guidelines for effective synthetic data generation,

F. K. Dankar and M. Ibrahim, “Fake it till you make it: Guidelines for effective synthetic data generation,”Applied Sciences, vol. 11, no. 5, p. 2158, 2021

2021
[15]

Generative ai in medicine and healthcare: promises, opportunities and challenges,

P. Zhang and M. N. Kamel Boulos, “Generative ai in medicine and healthcare: promises, opportunities and challenges,”Future Internet, vol. 15, no. 9, p. 286, 2023

2023
[16]

A medical chatbot for tunisian dialect using a rule-based and machine learning approach,

A. Rekik, M. Chebbi, and N. Boughanmi, “A medical chatbot for tunisian dialect using a rule-based and machine learning approach,” Journal of Medical Informatics, vol. 12, no. 3, pp. 45–56, 2023

2023
[17]

Towards building a chatbot-based first aid service in arabic language,

J. Anwar, P. Nadi, and N. Seddik, “Towards building a chatbot-based first aid service in arabic language,”Journal of Advanced Research in Applied Sciences and Engineering Technology, vol. 45, no. 2, pp. 1–10, 2025

2025
[18]

Pharmago: An online phar- maceutical ordering platform,

R. G. Gamage, K. U. Senadeera, N. S. Bandara, K. Y . Abeywardena, D. D. Diyamullage, and N. Amarasena, “Pharmago: An online phar- maceutical ordering platform,” in2021 3rd International Conference on Advancements in Computing (ICAC), 2021, pp. 364–370

2021
[19]

Lmrpa: Large lan- guage model-driven efficient robotic process automation for ocr,

O. H. Abdellaif, A. Nader, and A. Hamdi, “Lmrpa: Large lan- guage model-driven efficient robotic process automation for ocr,”arXiv preprint arXiv:2412.18063, 2024

2024 arXiv
[20]

Llm-sem: A sentiment- based student engagement metric using llms for e-learning platforms,

A. Hamdi, A. A. Mazrou, and M. Shaltout, “Llm-sem: A sentiment- based student engagement metric using llms for e-learning platforms,” arXiv preprint arXiv:2412.13765, 2024

2024 arXiv
[21]

Using ai to automatically process data from unstructured health records of patients with lung cancer,

M. Aldea, P. Rolland, S. Simon, A. Poplu, M. Wartelle, B. Vignal, J.-C. Louis, F. Lion, A. Borie, D. Planchardet al., “Using ai to automatically process data from unstructured health records of patients with lung cancer,”Cancer Research, vol. 84, no. 6 Supplement, pp. 3569–3569, 2024

2024
[22]

Riro: Reshaping inputs, refining outputs unlocking the potential of large language models in data-scarce contexts,

A. Hamdi, H. Kassab, M. Bahaa, and M. Mohamed, “Riro: Reshaping inputs, refining outputs unlocking the potential of large language models in data-scarce contexts,”arXiv preprint arXiv:2412.15254, 2024

2024 arXiv
[23]

Erpa: Efficient rpa model integrating ocr and llms for intelligent document processing,

O. H. Abdellaif, A. N. Hassan, and A. Hamdi, “Erpa: Efficient rpa model integrating ocr and llms for intelligent document processing,” in2024 International Mobile, Intelligent, and Ubiquitous Computing Conference (MIUCC). IEEE, 2024, pp. 295–300

2024

Pith tools

Reviewed August 4, 2026 · model on record in the stance chip above.

[1] [1]

D. E. Detmer, E. B. Steen, and R. S. Dick,The computer-based patient record: an essential technology for health care. National Academy Press, 1997

1997

[2] [2]

The concept of value in resource allocation,

T. C. Brown, “The concept of value in resource allocation,”Land economics, vol. 60, no. 3, pp. 231–246, 1984

1984

[3] [3]

Ai-assisted diagnosis and treatment plan- ning—a discussion of how ai can assist healthcare professionals in making more accurate diagnoses and treatment plans for diseases,

M. Rashid and M. Sharma, “Ai-assisted diagnosis and treatment plan- ning—a discussion of how ai can assist healthcare professionals in making more accurate diagnoses and treatment plans for diseases,” inAI in Disease Detection: Advancements and Applications, 2025, pp. 313– 336

2025

[4] [4]

Artificial intel- ligence and machine learning,

N. K ¨uhl, M. Schemmer, M. Goutier, and G. Satzger, “Artificial intel- ligence and machine learning,”Electronic Markets, vol. 32, no. 4, pp. 2235–2244, 2022

2022

[5] [5]

Hospital management system with chatbot,

M. L. PRIY ANKA, B. T. R. NAY AKA, B. S. RAKSHITHA, B. V AMSHI, and B. M. K. Y ADA V , “Hospital management system with chatbot,”International Journal of Mechanical Engineering Research and Technology, vol. 16, no. 2, pp. 144–154, 2024

2024

[6] [6]

Large language models in healthcare and medical domain: A review,

Z. A. Nazi and W. Peng, “Large language models in healthcare and medical domain: A review,”Informatics, vol. 11, no. 3, p. 57, 2024

2024

[7] [7]

A study of generative large language model for medical research and healthcare,

C. Peng, X. Yang, A. Chen, K. E. Smith, N. PourNejatian, A. B. Costa, C. Martin, M. G. Flores, Y . Zhang, T. Magocet al., “A study of generative large language model for medical research and healthcare,” NPJ digital medicine, vol. 6, no. 1, p. 210, 2023

2023

[8] [8]

Deep learning for arabic healthcare: Medicalbot,

M. Abdelhay, A. Mohammed, and H. A. Hefny, “Deep learning for arabic healthcare: Medicalbot,”Social Network Analysis and Mining, vol. 13, no. 1, p. 71, 2023

2023