Introduction: Provides an overview of the project and a short dicsussion on the

Introduction: Provides an overview of the project and a short dicsussion on the pertinent literature
Research question and methodology: Provides a clear statement on the goals of the project, an overview of the proposed approach, and a formal definition of the problem
Experimental results: Provides an overview of the dataset used for experiments, the metrics used for evaluating performances, and the experimental methodology. Presents experimental results as plots and/or tables
Concluding remarks: Provides a critical discussion on the experimental results and some ideas for future work. I need you to implement a study on python using google cola on this project: Bio Linking (P9)
Instructor: Darya Shlyk, Department of Computer Science, Università degli Studi di MilanoMuch of the world’s healthcare data is stored in free-text documents, such as medical records and clinical notes. Analyzing and extracting meaningful insights from this unstructured data can be challenging. One approach for extracting structured knowledge from a vast corpus of biomedical literature is to annotate the text with concepts from existing knowledge bases.In the context of biomedical information extraction, Concept Recognition (CR) is defined as a two-step process, that consists in identifying and linking textual mentions of biomedical entities, such as genes, diseases, and phenotypes, to corresponding concepts in domain-specific ontologies. The objective of this project is to devise effective strategies for automating the extraction of concepts from unstructured biomedical resources. When developing your CR system, you may decide to focus on a specific class of entities such as diseases or genes and link to any target ontology within the biomedical domain, such as MONDO, Human Phenotype Ontology, etc.For clinical CR, SNOMED CT (Systematized Nomenclature of Medicine Clinical Terms) is a viable option. SNOMED CT is a systematically organized clinical terminology that encompasses a wide range of clinical concepts, including diseases, symptoms, procedures, and body structures.DatasetPossible datasets include, PubMed abstracts and Clinical MIMIC-IV-Note available on request.ReferencesVuokko, Riikka, Anne Vakkuri, and Sari Palojoki. “Systematized Nomenclature of Medicine–Clinical Terminology (SNOMED CT) Clinical Use Cases in the Context of Electronic Health Record Systems: Systematic Literature Review.” JMIR medical informatics 11 (2023): e43750.
Lossio-Ventura, Juan Antonio, et al. “Clinical concept recognition: Evaluation of existing systems on EHRs.” Frontiers in Artificial Intelligence 5 (2023): 1051724.
Toonsi, Sumyyah, Şenay Kafkas, and Robert Hoehndorf. “BORD: A Biomedical Ontology based method for concept Recognition using Distant supervision: Application to Phenotypes and Diseases.” bioRxiv (2023): 2023-02.
Hailu, Negacy D., et al. “Biomedical concept recognition using deep neural sequence m
This is the GitHub link of the course, you can find additional informations about the models you have to use to solve this project: https://github.com/afflint/textsent/tree/master/2023-24