A machine learning framework for auto classification of imaging system exams in hospital setting for utilization optimization

Abstract

In clinical environment, Interventional X-Ray (IXR) system is used on various anatomies and for various types of the procedures. It is important to classify correctly each exam of IXR system into respective procedures and/or assign to correct anatomy. This classification enhances productivity of the system in terms of better scheduling of the Cath lab, also provides means to perform device usage/revenue forecast of the system by hospital management and focus on targeted treatment planning for a disease/anatomy. Although it may appear classification of each exam into respective procedure/anatomy a simple task. However, in real-life hospital settings, it is well-known that same system settings are used to perform different types of procedures. Though, such usage leads to under-utilization of the system. In this work, a method is developed to classify exams into respective anatomical type by applying machine-learning techniques (SVM, KNN and decision trees) on log information of the systems. The classification result is promising with accuracy of greater than 90%.

Automated Billing Code Retrieval from MRI Scanner Log Data

“…Moreover, the presented work may also be transferred to other medical (imaging) modalities by utilizing modality event logs to predict billing codes for billing purposes. For instance, the examined anatomy of interventional X-ray exams was already successfully retrieved from modality log data in [13]. Extending this work by training a prediction model with corresponding billing code data and tailoring the feature extraction process to this imaging modality may enable automated procedure billing coding for interventional X-ray systems.…”

“…Event logs from imaging modalities were used in [12] to retrieve the examined body region in MRI exams from sequence parameters with a classification accuracy of 94.7% and in [13] to classify interventional X-ray exams into respective procedures or examined anatomy, reaching a classification accuracy of 92.7%.…”

“…To summarize, it can be stated that a large part of the research has focused on the prediction of diagnosis codes [5–10] rather than on procedure billing codes [11]. Additionally, the data basis of the published methods for predicting medical codes usually consisted of discharge summaries and free-text reports, although there were recent attempts to utilize MRI log data for data analytic purposes [12, 13]. Thus, this paper is one of few works on automation of procedure coding, and to the best of our knowledge, it is the first attempt to focus on the prediction of procedure billing codes from data excluding EMRs or EHRs.…”

Aprendizaje Automático en Aplicaciones Fisioterapéuticas

“…En un ambiente clínico, los sistemas de intervención basados en rayos X (IXR, por sus siglas en inglés), son empleados para diferentes procedimientos en cardiología, neurología y angiología, en los cuales es sumamente importante clasificar correctamente cada examen del sistema IXR en los procedimientos respectivos y/o asignar la anatomía correcta, lo que representa un grado de oportunidad para la aplicación del aprendizaje máquina. Dicha clasificación mejora la calendarización y planeación de los tratamientos (Patil et al, 2016). En este trabajo se obtiene una exactitud mayor del 90% para la clasificación de exámenes, usando arboles de decisión, máquinas de soporte vectorial y k-vecinos más cercanos como plataformas de aprendizaje.…”