Automated sleep stage scoring employing a reasoning mechanism and evaluation of its explainability

Abstract

Scoring sleep stages from biological signals is an essential but labor-intensive inspection for sleep diagnosis. The existing automated scoring methods have achieved high accuracy but are not widely applied in clinical practice. In our understanding, the existing methods have failed to establish the trust of sleep experts (e.g., physicians and clinical technologists) due to a lack of ability to explain the evidences/clues for scoring. In this study, we developed a deep-learning-based scoring model with a reasoning mechanism called class activation mapping (CAM) to solve this problem. This mechanism explicitly shows which portions of the signals support our model’s sleep stage decision, and we verified that these portions overlap with the “characteristic waves,” which are evidences/clues used in the manual scoring process. In exchange for the acquisition of explainability, employing CAM makes it difficult to follow some scoring rules. Although we concerned the negative effect of CAM on the scoring accuracy, we have found that the impact is limited. The evaluation experiment shows that the proposed model achieved a scoring accuracy of $$86.9\%$$ 86.9 % . It is superior to those of some existing methods and the inter-rater reliability among the sleep experts. These results suggest that Sleep-CAM achieved both explainability and required scoring accuracy for practical usage.

Validation of sleep-staging accuracy for an in-home sleep electroencephalography device compared with simultaneous polysomnography in patients with obstructive sleep apnea

“…The portable EEG device used in this study uploads the EDF data to the cloud as soon as the overnight measurement is complete. Users can promptly review the previous night’s sleep results upon waking up, utilizing machine learning and AI to determine their sleep stages 33 , 34 . In this study, a professional EEG analyst performed all assessments to ensure reliability.…”

Validation of sleep-staging accuracy for an in-home sleep electroencephalography device: A comparison with simultaneous polysomnography in patients with obstructive sleep apnea

“…The portable EEG device used in this study uploads the EDF data to the cloud as soon as the overnight measurement is complete. Users can promptly review the previous night's sleep results upon waking up, utilizing machine learning and AI to determine their sleep stages [32,33]. In this study, a professional EEG analyst performed all assessments to ensure reliability.…”

Automatic sleep staging by a hybrid model based on deep 1D-ResNet-SE and LSTM with single-channel raw EEG signals

“…Moreover, exploring alternative loss functions and architectures may further improve the accuracy of sleep staging. Furthermore, the model’s interpretability needs to be explored to comprehend how it arrives at its classification decisions and identify potential errors or biases ( Phan et al, 2022 ; Jany et al, 2022 ; Horie et al, 2022 ). Meanwhile, addressing the relatively low performance for the N1 stage could be an area of focus for future research.…”