PURPOSE: Pneumonia is a common clinical diagnosis for which chest radiographs are often an important part of the diagnostic workup. Deep learning has the potential to expedite and improve the clinical interpretation of chest radiographs. While earlier approaches have emphasized the feasibility of "binary classification" to accomplish this task, alternative strategies may be possible. We explore the feasibility of a "semantic segmentation" deep learning approach to highlight the potential foci of pneumonia on frontal chest radiographs. MATERIALS AND METHODS: In this retrospective study, we trained a U-net convolutional neural network (CNN) to predict pixel-wise probability maps for pneumonia using a public data set provided by the Radiological Society of North America (RSNA) comprised of 22,000 radiographs and radiologist-defined bounding boxes. We reserved 3684 radiographs as an independent validation data set and assessed overall performance for localization using Dice overlap and classification performance using the area under the receiver-operator characteristic curve. RESULTS: For classification/detection of pneumonia, area under the receiver-operator characteristic curve on frontal radiographs was 0.854 with a sensitivity of 82.8% and specificity of 72.6%. Using this strategy of neural network training, probability maps localized pneumonia to lung parenchyma for essentially all validation cases. For segmentation of pneumonia for positive cases, predicted probability maps had a mean Dice score (±SD) of 0.603±0.204, and 60.0% of these had a Dice score >0.5. CONCLUSIONS: A "semantic segmentation" deep learning approach can provide a probabilistic map to assist in the diagnosis of pneumonia. In combination with the patient's history, clinical findings and other imaging, this strategy may help expedite and improve diagnosis.
PURPOSE: Pneumonia is a common clinical diagnosis for which chest radiographs are often an important part of the diagnostic workup. Deep learning has the potential to expedite and improve the clinical interpretation of chest radiographs. While earlier approaches have emphasized the feasibility of "binary classification" to accomplish this task, alternative strategies may be possible. We explore the feasibility of a "semantic segmentation" deep learning approach to highlight the potential foci of pneumonia on frontal chest radiographs. MATERIALS AND METHODS: In this retrospective study, we trained a U-net convolutional neural network (CNN) to predict pixel-wise probability maps for pneumonia using a public data set provided by the Radiological Society of North America (RSNA) comprised of 22,000 radiographs and radiologist-defined bounding boxes. We reserved 3684 radiographs as an independent validation data set and assessed overall performance for localization using Dice overlap and classification performance using the area under the receiver-operator characteristic curve. RESULTS: For classification/detection of pneumonia, area under the receiver-operator characteristic curve on frontal radiographs was 0.854 with a sensitivity of 82.8% and specificity of 72.6%. Using this strategy of neural network training, probability maps localized pneumonia to lung parenchyma for essentially all validation cases. For segmentation of pneumonia for positive cases, predicted probability maps had a mean Dice score (±SD) of 0.603±0.204, and 60.0% of these had a Dice score >0.5. CONCLUSIONS: A "semantic segmentation" deep learning approach can provide a probabilistic map to assist in the diagnosis of pneumonia. In combination with the patient's history, clinical findings and other imaging, this strategy may help expedite and improve diagnosis.
Authors: Kang Wang; Adrija Mamidipalli; Tara Retson; Naeim Bahrami; Kyle Hasenstab; Kevin Blansit; Emily Bass; Timoteo Delgado; Guilherme Cunha; Michael S Middleton; Rohit Loomba; Brent A Neuschwander-Tetri; Claude B Sirlin; Albert Hsiao Journal: Radiol Artif Intell Date: 2019-03-27
Authors: Brian Hurt; Meagan A Rubel; Evan M Masutani; Kathleen Jacobs; Lewis Hahn; Michael Horowitz; Seth Kligerman; Albert Hsiao Journal: J Thorac Imaging Date: 2022-03-01 Impact factor: 5.528
Authors: Jeremy A Irvin; Anuj Pareek; Jin Long; Pranav Rajpurkar; David Ken-Ming Eng; Nishith Khandwala; Peter J Haug; Al Jephson; Karen E Conner; Benjamin H Gordon; Fernando Rodriguez; Andrew Y Ng; Matthew P Lungren; Nathan C Dean Journal: J Thorac Imaging Date: 2021-09-23 Impact factor: 5.528
Authors: Morgan Carlile; Brian Hurt; Albert Hsiao; Michael Hogarth; Christopher A Longhurst; Christian Dameff Journal: J Am Coll Emerg Physicians Open Date: 2020-11-05