LaboRecommender: A crazy-easy to use Python-based recommender system for laboratory tests
This addresses the challenge for physicians in efficiently navigating complex laboratory test options, though it is incremental as it applies an existing method to a new domain.
The paper tackled the problem of selecting appropriate laboratory tests from a rapidly expanding set by developing a recommender system using collaborative filtering, which achieved 95.54% mean average precision.
Laboratory tests play a major role in clinical decision making because they are essential for the confirmation of diagnostics suspicions and influence medical decisions. The number of different laboratory tests available to physicians in our age has been expanding very rapidly due to the rapid advances in laboratory technology. To find the correct desired tests within this expanding plethora of elements, the Health Information System must provide a powerful search engine and the practitioner need to remember the exact name of the laboratory test to correctly select the bag of tests to order. Recommender systems are platforms which suggest appropriate items to a user after learning the users' behaviour. A neighbourhood-based collaborative filtering method was used to model the recommender system, where similar bags, clustered using nearest neighbours algorithm, are used to make recommendations of tests for each other similar bag of laboratory tests. The recommender system developed in this paper achieved 95.54 % in the mean average precision metric. A fully documented Python package named LaboRecommender was developed to implement the algorithm proposed in this paper