A Deep Learning Framework for Automated ICD-10 Coding.

The International Statistical Classification of Diseases and Related Health Problems (ICD) is one of the widely used classification system for diagnoses and procedures to assign diagnosis codes to Electronic Health Record (EHR) associated with a patient's stay. The aim of this paper is to propose an automated coding system to assist physicians in the assignment of ICD codes to EHR. For this purpose, we created a pipeline of Natural Language Processing (NLP) and Deep Learning (DL) models able to extract the useful information from French medical texts and to perform classification. After the evaluation phase, our approach was able to predict 346 diagnosis codes from heterogeneous medical units with an accuracy average of 83%. Our results were finally validated by physicians of the Medical Information Department (MID) in charge of coding hospital stays.

The detection of hospitalized patients at risk of testing positive to multi-drug resistant bacteria using MOCA-I, a rule-based "white-box" classification algorithm for medical data.

BACKGROUND: Multi-drug resistant (MDR) bacteria are a major health concern. In this retrospective study, a rule-based classification algorithm, MOCA-I (Multi-Objective Classification Algorithm for Imbalanced data) is used to identify hospitalized patients at risk of testing positive for multidrug-resistant (MDR) bacteria, including Methicillin-resistant Staphylococcus aureus (MRSA), before or during their stay. METHODS: Applied to a data set of 48,945 hospital stays (including known cases of carriage) with up to 16,325 attributes per stay, MOCA-I generated alert rules for risk of carriage or infection. A risk score was then computed from each stay according to the triggered rules.Recall and precision curves were plotted. RESULTS: The classification can be focused on specifically detecting high risk of having a positive test, or identifying large numbers of at-risk patients by modulating the risk score cut-off level. For a risk score above 0.85,recall (sensitivity) is 62 % with 69 % precision (confidence) for MDR bacteria, recall is 58 % with 88 % precision for MRSA. In addition, MOCA-I identifies 38 and 21 cases of previously unknown MDR and MRSA respectively. CONCLUSIONS: MOCA-I generates medically pertinent alert rules. This classification algorithm can be used to detect patients with high risk of testing positive to MDR bacteria (including MRSA). Classification can be modulated by appropriately setting the risk score cut-off level to favor specific detection of small numbers of patients at very high risk or identification of large numbers of patients at risk. MOCA-I can thus contribute to more adapted treatments and preventive measures from admission, depending on the clinical setting or management strategy.

PoleSat's Modelling a Reorganization of Vascular Catheterization Provision at the Territorial Hospital Grouping Level.

Since a French organization (2016) has defined "the territorial hospital groupings", public hospitals must share medical-economic knowledge and decision-makers expect prospective analyses. PoleSat aims, quick hospital-catchment area modellings, completed by population analyses. Modellings are based on "diagnostic and interventional vascular catheterizations" acts and Nouvelle-Aquitaine, and they are carried out 3 times, through the graphical user interface's main-setting values, coupled with 3 activity-scenarios. Scenario results cannot confirm the NA02-Atlantique's H0. The experts have approved PoleSat's method as a robust help-tool; therefore they project to repeat its usages.

Polesat-Web-2018: A Simulation IT Tool with Immediate Prospective and Strategic Views of Hospital Spatial Planning.

Medical geo-informatics allows the Health world to address major challenges thanks to attractive concepts, methods and user-friendly IT. PoleSat-web-2018 presents a decision support system - a modelling "variable geometry" IT tool for simulation of hospital spatial planning. The outputs enable quasi-instantaneous analytic visualization at several geographic levels. PoleSat-web-2018 provides prospective views of hospital catchments (by grouping, closing) and proves to be relevant for the French planners of the Ministry of Health.

EHR Usage Problems: A Preliminary Study.

Electronic health record (EHR) systems were initially developed to improve health care delivery by facilitating the healthcare professionals' access to electronically-stored patient information, but problems are regularly reported in the literature. We present here a preliminary study conducted at a 950-bed university hospital. They have implemented an EHR in 2012 to remove their paper-based system. After few years, physicians complain that the EHR is "too complex", "too slow", "unsatisfying", and "which interacts with too many health software". This preliminary study was based on individual interviews inspired from critical incident technique with 9 hospital professionals (physicians and pharmacist) to establish a global diagnostic of the EHR's usability failures/difficulties and their potential impacts. Results show that professionals faced to many constraints impacting their work but more importantly the patient care, with recent outstanding examples. This work is a first step of a larger study to help the hospital to map usability failures, their context of use and associated risks/impacts, and to provide solutions to fix it.