Hossein Rouhizadeh

Hossein Rouhizadeh

Auto-generated profile, based on publicly available information

Hossein Rouhizadeh

Department of Radiology and Medical Informatics, Faculty of Medicine, University of Geneva, Geneva, Switzerland

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
4 May 2024 in Scientific Data
Verified
Journal Article
Open access
PDF file
HRHossein Rouhizadeh
Hossein Rouhizadeh
INIrina Nikishina
Irina Nikishina
AYAnthony Yazdani
Anthony Yazdani
ABAlban Bornet
Alban Bornet
BZBoya Zhang
Boya Zhang
DTDouglas Teodoro
Douglas Teodoro
Due to the complexity of the biomedical domain, the ability to capture semantically meaningful representations of terms in context is a long-standing challenge. Despite important progress in the past years, no evaluation benchmark has been developed to evaluate how well language models represent biomedical concepts according to their corresponding context. Inspired by the Word-in-Context (WiC) benchmark, in which word sense disambiguation is reformulated as a binary classification task, we propose a novel dataset, BioWiC, to evaluate the ability of language models to encode biomedical terms in context. BioWiC comprises 20’156 instances, covering over 7’400 unique biomedical terms, making it the largest WiC dataset in the biomedical domain. We evaluate BioWiC both intrinsically and extrinsically and show that it could be used as a reliable benchmark for evaluating context-dependent embeddings in biomedical corpora. In addition, we conduct several experiments using a variety of discriminative and generative large language models to establish robust baselines that can serve as a foundation for future research.

15 December 2025 in Research Square
Verified
Preprint
Open access
PDF file
JHJiawei He
Jiawei He
BZBoya Zhang
Boya Zhang
HRHossein Rouhizadeh
Hossein Rouhizadeh
YCYingjian Chen
Yingjian Chen
RYRui Yang
Rui Yang
DTDouglas Teodoro
Douglas Teodoro
Large language models (LLMs) in biomedicine face a fundamental conflict between static parameter knowledge and the dynamic nature of clinical evidence. Retrieval-Augmented Generation (RAG) addresses this by grounding generation in external data, yet it introduces new complexities in latency and architecture. This survey synthesizes the biomedical RAG landscape (2020–2025), classifying systems into naive, advanced, and modular paradigms. Beyond a technological taxonomy, we formalize the biomedical RAG trilemma, identifying the inherent trade-offs between reasoning depth, inference latency, and data privacy that constrain current clinical deployment. We analyze how recent agentic workflows enhance diagnostic reasoning but risk prohibitive latency, and how privacy constraints dictate the choice between powerful cloud-based models and local deployment. Finally, we outline the alignment gap in multimodal RAG and propose future directions for self-correcting, verifiable clinical agents.

15 January 2025 in medRxiv
Verified
Preprint
Open access
PDF file
HRHossein Rouhizadeh
Hossein Rouhizadeh
AYAnthony Yazdani
Anthony Yazdani
BZBoya Zhang
Boya Zhang
DADavid Vicente Alvarez
David Vicente Alvarez
MHMatthias Hueser
Matthias Hueser
DTDouglas Teodoro
Douglas Teodoro
Interoperability in health information systems is crucial for accurate data exchange across environments such as electronic health records, clinical notes, and medical research. The main challenge arises from the wide variation in biomedical concepts, their representation across different systems and languages, and the limited context, complicating data integration and standardization. Inspired by recent advances in large language models (LLMs), this study explores their potential role as biomedical knowledge engineers to (semi-)automate multilingual biomedical concept normalization, a key task for semantic interoperability of medical concepts. We developed a novel multilingual dataset comprising 59’104 unique terms mapped to 27’280 distinct biomedical concepts, designed to assess language model performance across this task within five European languages: English, French, German, Spanish, and Turkish. We then proposed a multi-stage pipeline based on a retrieve-then-rerank approach using sparse and dense retrievers, rerankers, and fusion approaches, leveraging discriminative and generative LLMs, with a predefined primary knowledge organization system. Our experiments show that the best discriminative model, e5, achieves an accuracy of 71%, surpassing the best generative model, Mistral, by 2% (p-value < 0.001). For semi-automated workflows, e5 maintained superior performance with 82% recall@10 versus Mistral’s 78%. Our findings demonstrate a pathway to how LLM-based approaches can advance the normalization of multilingual biomedical terms as well as the limitations of LLMs in encoding biomedical concepts.

29 October 2020
Verified
Conference Paper
Closed access
HRHossein Rouhizadeh
Hossein Rouhizadeh
MSMehrnoush Shamsfard
Mehrnoush Shamsfard
MRMasoud Rouhizadeh
Masoud Rouhizadeh
Word Sense Disambiguation (WSD) can be the key component of downstream NLP applications. Existing WSD methods and systems are mostly developed and evaluated on English and low-resource languages such as Persian have not been well studied. In this paper, we propose a new knowledge-based method for Persian WSD. Using a pre-trained LDA model, we retrieve the topics of each document and assign each ambiguous content word to one of the topics. For each possible sense s of a given word w, we compute the similarity between the FarsNet (the Persian WordNet) gloss of s and the words of the assigned topic of w. We then choose the sense with the highest score as the most probable one. We evaluated our method on a Persian all-words WSD dataset and show that, compared to other knowledge-based methods, we could achieve state-of-the-art performance.

15 May 2025
Verified
Book Chapter
Open access
PDF file
AYAnthony Yazdani
Anthony Yazdani
HRHossein Rouhizadeh
Hossein Rouhizadeh
Alban BornetAlban Bornet
Alban Bornet
DTDouglas Teodoro
Douglas Teodoro
Adverse drug event (ADE) detection in social media texts poses significant challenges due to the informal nature of the text and the limited availability of annotations. The scarcity of ADE named entity recognition (NER) datasets for social media hinders the development of robust ADE detection models for this type of corpus. In this paper, we leveraged the generative capabilities of large language models (LLMs) to create synthetic data, addressing this dataset gap. Specifically, we generated 17,000 tweets with ADE annotations and pre-trained NER models on this synthetic data. Our evaluations on an out-of-sample collection of 915 manually annotated tweets revealed that these models outperform state-of-the-art lexico-based and massively pre-trained open NER models. We also show that fine-tuning our synthetically pre-trained models on human-annotated data surpasses the current state-of-the-art in ADE detection on tweets. These findings suggest that synthetic data generated by LLMs can enhance ADE detection performance, offering a promising avenue to explore in response to the scarcity of annotated ADE datasets. The synthetic dataset is available at https://huggingface.co/datasets/anthonyyazdaniml/synthetic-ner-ade-tweets-v1.

8 May 2026 in Journal of the American Medical Informatics Association
Verified
Journal Article
Open access
HRHossein Rouhizadeh
Hossein Rouhizadeh
ASAbiram Sandralegar
Abiram Sandralegar
AYAnthony Yazdani
Anthony Yazdani
WFWeibo Feng
Weibo Feng
OSOren Schreier
Oren Schreier
DTDouglas Teodoro
Douglas Teodoro
Objectives: The automation of medical report generation using large language models (LLMs) could significantly reduce physicians’ documentation burden while enhancing healthcare efficiency. However, the misuse of generative artificial intelligence in medical reporting can lead to important safety risks for patients. We addressed 2 questions: (1) What is the quality of medical reports generated by LLMs in English and French? and (2) Can we distinguish between human-written and LLM-generated medical reports? Materials and methods: We evaluated the quality of reports generated by several multilingual, open-weight LLMs using text similarity metrics on 4212 medical reports in English and French across multiple specialties. A bilingual expert panel of certified physicians (n = 4) and medical residents (n = 5) scored accuracy, fluency, and completeness of generated reports using a 1-5 Likert scale. Experts also completed a Turing-like test, blindly identifying reports as human or machine-generated. Results: Phi-4 achieved the best overall performance (ROUGE-1: 0.70, BERTScore: 0.83). Expert evaluation confirmed high-quality reports in both languages (overall 4.6/5.0). Medical experts performed better than chance but struggled to differentiate human versus machine reports (accuracy: 0.60). Automatic classifiers showed strong performance (accuracy: 0.98). Discussion: The high quality of LLM-generated reports supports their potential to enhance healthcare efficiency in multilingual settings. However, the discrepancy between human detection difficulty and automated detection success reveals inherent limitations in relying solely on human oversight for quality assurance and misuse prevention. Conclusions: Deployment of LLMs for medical reporting requires combining automated detection tools with human expertise to ensure patient safety. Dataset and code: https://github.com/ds4dh/medical_report_generation.