Chiao-Chin Lee1, Chin-Sheng Lin1, Chien-Sung Tsai2, Tien-Ping Tsao3, Cheng-Chung Cheng1, Jun-Ting Liou1, Wei-Shiang Lin1, Chia-Cheng Lee4,5, Jiann-Torng Chen6, Chin Lin7,8,9. 1. Division of Cardiology, Department of Internal Medicine, Tri-Service General Hospital, National Defense Medical Center, Taipei, Taiwan, ROC. 2. Division of Cardiovascular Surgery, Department of Surgery, Tri-Service General Hospital, National Defense Medical Center, Taipei, Taiwan, ROC. 3. Department of Cardiology, Cheng Hsin Hospital, Taipei, Taiwan, ROC. 4. Planning and Management Office, Tri-Service General Hospital, National Defense Medical Center, Taipei, Taiwan, ROC. 5. Division of Colorectal Surgery, Department of Surgery, Tri-Service General Hospital, National Defense Medical Center, Taipei, Taiwan, ROC. 6. Department of Ophthalmology, Tri-Service General Hospital, National Defense Medical Center, Taipei, Taiwan, ROC. 7. Graduate Institute of Life Sciences, National Defense Medical Center, No.161, Min-Chun E. Rd., Sec. 6, Neihu, Taipei 114, Taiwan, ROC. xup6fup0629@gmail.com. 8. School of Medicine, National Defense Medical Center, Taipei, Taiwan, ROC. xup6fup0629@gmail.com. 9. School of Public Health, National Defense Medical Center, Taipei, Taiwan, ROC. xup6fup0629@gmail.com.
Abstract
PURPOSE: To determine if an electrocardiogram-based artificial intelligence system can identify pneumothorax prior to radiological examination. METHODS: This is a single-center, retrospective, electrocardiogram-based artificial intelligence (AI) system study that included 107 ECGs from 98 pneumothorax patients. Seven patients received needle decompression due to tension pneumothorax, and the others received thoracostomy due to instability (respiratory rate ≥ 24 breaths/min; heart rate, < 60 beats/min or > 120 beats/min; hypotension; room air O2 saturation, < 90%; and patient could not speak in whole sentences between breaths). Traumatic pneumothorax and bilateral pneumothorax were excluded. The ECGs of 132,127 patients presenting to the emergency department without pneumothorax were used as the control group. The development cohort included approximately 80% of the ECGs for training the deep learning model (DLM), and the other 20% of ECGs were used to validate the performance. A human-machine competition involving three physicians was conducted to assess the model performance. RESULTS: The areas under the receiver operating characteristic (ROC) curves (AUCs) of the DLM in the validation cohort and competition set were 0.947 and 0.957, respectively. The sensitivity and specificity of our DLM were 94.7% and 88.1% in the validation cohort, respectively, which were significantly higher than those of all physicians. Our DLM could also recognize the location of pneumothorax with 100% accuracy. Lead-specific analysis showed that lead I ECG made a major contribution, achieving an AUC of 0.930 (94.7% sensitivity, 86.0% specificity). The inclusion of the patient characteristics allowed our AI system to achieve an AUC of 0.994. CONCLUSION: The present AI system may assist the medical system in the early identification of pneumothorax through 12-lead ECG, and it performs as well with lead I ECG alone as with 12-lead ECG.
PURPOSE: To determine if an electrocardiogram-based artificial intelligence system can identify pneumothorax prior to radiological examination. METHODS: This is a single-center, retrospective, electrocardiogram-based artificial intelligence (AI) system study that included 107 ECGs from 98 pneumothorax patients. Seven patients received needle decompression due to tension pneumothorax, and the others received thoracostomy due to instability (respiratory rate ≥ 24 breaths/min; heart rate, < 60 beats/min or > 120 beats/min; hypotension; room air O2 saturation, < 90%; and patient could not speak in whole sentences between breaths). Traumatic pneumothorax and bilateral pneumothorax were excluded. The ECGs of 132,127 patients presenting to the emergency department without pneumothorax were used as the control group. The development cohort included approximately 80% of the ECGs for training the deep learning model (DLM), and the other 20% of ECGs were used to validate the performance. A human-machine competition involving three physicians was conducted to assess the model performance. RESULTS: The areas under the receiver operating characteristic (ROC) curves (AUCs) of the DLM in the validation cohort and competition set were 0.947 and 0.957, respectively. The sensitivity and specificity of our DLM were 94.7% and 88.1% in the validation cohort, respectively, which were significantly higher than those of all physicians. Our DLM could also recognize the location of pneumothorax with 100% accuracy. Lead-specific analysis showed that lead I ECG made a major contribution, achieving an AUC of 0.930 (94.7% sensitivity, 86.0% specificity). The inclusion of the patient characteristics allowed our AI system to achieve an AUC of 0.994. CONCLUSION: The present AI system may assist the medical system in the early identification of pneumothorax through 12-lead ECG, and it performs as well with lead I ECG alone as with 12-lead ECG.