Abstract：
OBJECTIVE: To develop and validate machine learning (ML) and deep learning (DL) models using drug-induced sleep endoscopy (DISE) images to predict the therapeutic efficacy of hypoglossal nerve stimulator (HGNS) implantation.
METHODS: Patients who underwent DISE and subsequent HGNS implantation at a tertiary care referral center were included. Six DL models and five ML algorithms were trained on images from the base of tongue (BOT) and velopharynx (VP) from patients classified as responders or non-responders as defined by Sher\'s criteria (50% reduction in apnea-hypopnea index (AHI) and AHI < 15 events/h). Precision, recall, F1 score, and overall accuracy were evaluated as measures of performance.
RESULTS: In total, 25,040 images from 127 patients were included, of which 16,515 (69.3%) were from responders and 8,262 (30.7%) from non-responders. Models trained on the VP dataset had greater overall accuracy when compared to BOT alone and combined VP and BOT image sets, suggesting that VP images contain discriminative features for identifying therapeutic efficacy. The VCG-16 DL model had the best overall performance on the VP image set with high training accuracy (0.833), F1 score (0.78), and recall (0.883). Among ML models, the logistic regression model had the greatest accuracy (0.685) and F1 score (0.813).
CONCLUSIONS: Deep neural networks have potential to predict HGNS therapeutic efficacy using images from DISE, facilitating better patient selection for implantation. Development of multi-institutional data and image sets will allow for development of generalizable predictive models.
METHODS: N/A Laryngoscope, 2024.
METHODS: Patients who underwent DISE and subsequent HGNS implantation at a tertiary care referral center were included. Six DL models and five ML algorithms were trained on images from the base of tongue (BOT) and velopharynx (VP) from patients classified as responders or non-responders as defined by Sher\'s criteria (50% reduction in apnea-hypopnea index (AHI) and AHI < 15 events/h). Precision, recall, F1 score, and overall accuracy were evaluated as measures of performance.
RESULTS: In total, 25,040 images from 127 patients were included, of which 16,515 (69.3%) were from responders and 8,262 (30.7%) from non-responders. Models trained on the VP dataset had greater overall accuracy when compared to BOT alone and combined VP and BOT image sets, suggesting that VP images contain discriminative features for identifying therapeutic efficacy. The VCG-16 DL model had the best overall performance on the VP image set with high training accuracy (0.833), F1 score (0.78), and recall (0.883). Among ML models, the logistic regression model had the greatest accuracy (0.685) and F1 score (0.813).
CONCLUSIONS: Deep neural networks have potential to predict HGNS therapeutic efficacy using images from DISE, facilitating better patient selection for implantation. Development of multi-institutional data and image sets will allow for development of generalizable predictive models.
METHODS: N/A Laryngoscope, 2024.
摘要：
目的：使用药物诱导睡眠内窥镜（DISE）图像开发和验证机器学习（ML）和深度学习（DL）模型，以预测舌下神经刺激器（HGNS）植入的治疗效果。
方法：纳入在三级护理转诊中心接受DISE和随后HGNS植入的患者。根据Sher标准（呼吸暂停低通气指数（AHI）和AHI<15事件/h）定义的分类为应答者或无应答者的患者的舌根（BOT）和velopharynx（VP）的图像，对六个DL模型和五个ML算法进行了训练。Precision,召回，F1得分,和总体准确性被评估为绩效指标。
结果：总计，包括来自127名患者的25,040张图像，其中16,515人(69.3%)来自应答者，8,262人(30.7%)来自非应答者。与单独的BOT以及组合的VP和BOT图像集相比，在VP数据集上训练的模型具有更高的总体准确性，这表明VP图像含有鉴别治疗效果的特征。VCG-16DL模型在VP图像集上具有最佳的整体性能，具有较高的训练精度（0.833），F1得分（0.78），和召回（0.883）。在ML模型中，logistic回归模型的准确度最高(0.685),F1评分最高(0.813).
结论：深度神经网络有可能使用来自DISE的图像来预测HGNS治疗效果，有利于更好的患者选择植入。多机构数据和图像集的开发将允许开发可推广的预测模型。
方法：N/A喉镜，2024.
方法：纳入在三级护理转诊中心接受DISE和随后HGNS植入的患者。根据Sher标准（呼吸暂停低通气指数（AHI）和AHI<15事件/h）定义的分类为应答者或无应答者的患者的舌根（BOT）和velopharynx（VP）的图像，对六个DL模型和五个ML算法进行了训练。Precision,召回，F1得分,和总体准确性被评估为绩效指标。
结果：总计，包括来自127名患者的25,040张图像，其中16,515人(69.3%)来自应答者，8,262人(30.7%)来自非应答者。与单独的BOT以及组合的VP和BOT图像集相比，在VP数据集上训练的模型具有更高的总体准确性，这表明VP图像含有鉴别治疗效果的特征。VCG-16DL模型在VP图像集上具有最佳的整体性能，具有较高的训练精度（0.833），F1得分（0.78），和召回（0.883）。在ML模型中，logistic回归模型的准确度最高(0.685),F1评分最高(0.813).
结论：深度神经网络有可能使用来自DISE的图像来预测HGNS治疗效果，有利于更好的患者选择植入。多机构数据和图像集的开发将允许开发可推广的预测模型。
方法：N/A喉镜，2024.
