Natural Language Explanatory Arguments for Correct and Incorrect Diagnoses of Clinical Cases

Marro, Santiago; Molinet, Benjamin; Cabrio, Elena; Villata, Serena

doi:10.5220/0011927000003393

Cited by 1 publication

(4 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The main contribution of the paper is twofold: first, we present two novel linguistic resources for the medical domain, i.e., the MEDQA-UMLS-Symp dataset which contains a set of clinical case descriptions together with a set of possible questions and answers on the correct diagnosis from MedQA [9], annotated with medical entities from UMLS [10], and a database of biological boundaries for common findings; second, we introduce a complete pipeline to generate natural language explanations for correct and incorrect diagnosis, relying on clinical entities detected from clinical cases and aligned with medical ontologies. This journal paper extends our previous contribution [11] showing that medical findings and vital signs can enhance explanations by converting observed values to a medical terminology, based on a manually verified medical database and large Language Models, and enriching explanations with findings information.…”

Section: Introductionsupporting

confidence: 61%

“…However, only a limited number of studies have focused on disease and medical findings annotation, e.g., the NCBI disease corpus [47] and the MEDQA-USMLE-Symp dataset [11], which is annotated with UMLS symptoms and findings tags. Despite these two resources, the issue of matching medical findings to symptoms is still an open research question.…”

Section: Medical Data and Linguistic Resourcesmentioning

confidence: 99%

“…Finally, Raza et al [58] propose the Bio-Epidemiology-Ner (BioEN) pipeline, an approach inspired by [11], where they fine-tune a DistilBERT [59] model, a simplified and more computationally efficient version of BERT, for the task of biomedical NER. They adapt the last layer of the pre-trained DistilBERT model to their specific biomedical task and adjust the input and output dimensions accordingly.…”

Section: Information Extraction On Medical Textmentioning

confidence: 99%

“…We first evaluate our semi-automatic approach on a similar database of laboratory reference ranges for healthy adults 11 , in order to see if such Large Language Model (LLM) would be useful for assisting the database creation. This selected database provides reference ranges for various categories such as Electrolytes, Hematology, Lipids, Acid base, Gastrointestinal function, Cardiac enzymes, Hormones, Vitamins, Tumor markers, and Miscellaneous, but does not specify the medical terms associated with these values.…”

Section: Automated Evaluationmentioning

confidence: 99%

See 3 more Smart Citations

Explanatory argumentation in natural language for correct and incorrect medical diagnoses

Molinet,

Marro,

Cabrio

et al. 2024

J Biomed Semant

Self Cite

View full text Add to dashboard Cite

Background A huge amount of research is carried out nowadays in Artificial Intelligence to propose automated ways to analyse medical data with the aim to support doctors in delivering medical diagnoses. However, a main issue of these approaches is the lack of transparency and interpretability of the achieved results, making it hard to employ such methods for educational purposes. It is therefore necessary to develop new frameworks to enhance explainability in these solutions. Results In this paper, we present a novel full pipeline to generate automatically natural language explanations for medical diagnoses. The proposed solution starts from a clinical case description associated with a list of correct and incorrect diagnoses and, through the extraction of the relevant symptoms and findings, enriches the information contained in the description with verified medical knowledge from an ontology. Finally, the system returns a pattern-based explanation in natural language which elucidates why the correct (incorrect) diagnosis is the correct (incorrect) one. The main contribution of the paper is twofold: first, we propose two novel linguistic resources for the medical domain (i.e, a dataset of 314 clinical cases annotated with the medical entities from UMLS, and a database of biological boundaries for common findings), and second, a full Information Extraction pipeline to extract symptoms and findings from the clinical cases and match them with the terms in a medical ontology and to the biological boundaries. An extensive evaluation of the proposed approach shows the our method outperforms comparable approaches. Conclusions Our goal is to offer AI-assisted educational support framework to form clinical residents to formulate sound and exhaustive explanations for their diagnoses to patients.

show abstract

Section: Introductionsupporting

confidence: 61%

Section: Medical Data and Linguistic Resourcesmentioning

confidence: 99%