Application on Reinforcement Learning for Diagnosis Based on Medical Image
By Stelmo Magalhães, Barros Netto, V. Rodrigues, C. Leite, Aristófanes Corrêa Silva, Anselmo Cardoso de Paiva, Areolino A. Neto. 2008
21 citations recorded by the source corpus.
The benefit of a medical imaging examination in terms of its ability to yield an accurate diagnosis depends on the quality of both the image acquisition and the image interpretation. During the past century, radiology has grown tremendously due to advances in image detector systems and computer technology. On the other side, the image interpretation is an error prone task. The number of lawsuits filed against medical imaging professionals that are related to the miss of a diagnosis is close to 70% (Berlin, 1995). The most common errors are perceptual errors that lead to diagnoses misses, representing about 60% of the cases (Renfrew et al., 1992). In cancer diagnoses, some studies show that the risks of false negative diagnoses are up to 75%. These number of diagnosis errors due to misinterpretation of medical images and the current development in automation methods to help the specialist in this error prone task, lead to the development of novel devices and techniques to assist the specialist in the diagnosis achievement. These systems can provide a second opinion and may be used as a first stage of radiological interpretation (Nab et al, 1992). These systems are commonly named Computer-Aided Diagnosis (CAD) systems and have been developed to assist radiologists and other specialized physicians in the diagnostic setting like early detection of lung cancer in radiographs and CT images. The system aid may contribute to understand essential features and information hidden in the images which are not readily apparent. Also, the effects between the different image aspects are not distinguishable. The image information and the extracted parameters may be too complex to be solved with conventional techniques. Thus we may use other techniques like Machine Learning methods. Machine Learning methods use these complex sets of data, and can help to model the nonlinear relationships that exist between them, improving medical care. Machine Learning (ML) aims at providing techniques and methods for accumulating, changing and updating knowledge in computational systems, and in particular mechanisms to help the system to induce knowledge from examples or new data. These
Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Radiology, Nuclear Medicine and Imaging
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
In Computer Vision and Pattern Recognition
In Radiology, Nuclear Medicine and Imaging
Citations in the corpus
- Reinforcement Learning in Healthcare: A Survey (2019, methodology): “Netto et al. [165] presented an overview of RL in medical image applications, providing a detailed illustration of particular use of RL for lung nodules classification.”
- Learning to Diagnose: Assimilating Clinical Narratives using Deep Reinforcement Learning (2017, methodology): “Prior works that use reinforcement learning for clinical decision support tasks focused on other modalities e.g. medical imaging (Netto et al., 2008) or specific domain-dependent use cases, and clinical trials (Poolla, 2003; Shortreed et al., 2011; Zhao et al., 2011), but not for inferencing…”
- Diagnostic Inferencing via Improving Clinical Concept Extraction with Deep Reinforcement Learning: A Preliminary Study (2017, methodology): “medical imaging (Netto et al., 2008) or specific domain-dependent use cases and clinical trials (Poolla, 2003; Shortreed et al.”