RESUMEN
Accurate and efficient detection of attention-deficit/hyperactivity disorder (ADHD) is critical to ensure proper treatment for affected individuals. Current clinical examinations, however, are inefficient and prone to misdiagnosis, as they rely on qualitative observations of perceived behavior. We propose a robust machine learning based framework that analyzes pupil-size dynamics as an objective biomarker for the automated detection of ADHD. Our framework integrates a comprehensive pupillometric feature engineering and visualization pipeline with state-of-the-art binary classification algorithms and univariate feature selection. The support vector machine classifier achieved an average 85.6% area under the receiver operating characteristic (AUROC), 77.3% sensitivity, and 75.3% specificity using ten-fold nested cross-validation (CV) on a declassified dataset of 50 patients. 218 of the 783 engineered features, including fourier transform metrics, absolute energy, consecutive quantile changes, approximate entropy, aggregated linear trends, as well as pupil-size dilation velocity, were found to be statistically significant differentiators (p < 0.05), and provide novel behavioral insights into associations between pupil-size dynamics and the presence of ADHD. Despite a limited sample size, the strong AUROC values highlight the robustness of the binary classifiers in detecting ADHD-as such, with additional data, sensitivity and specificity metrics can be substantially augmented. This study is the first to apply machine learning based methods for the detection of ADHD using solely pupillometrics, and highlights its strength as a potential discriminative biomarker, paving the path for the development of novel diagnostic applications to aid in the detection of ADHD using oculometric paradigms and machine learning.
Asunto(s)
Trastorno por Déficit de Atención con Hiperactividad/diagnóstico , Trastorno por Déficit de Atención con Hiperactividad/metabolismo , Biomarcadores/metabolismo , Algoritmos , Niño , Femenino , Humanos , Aprendizaje Automático , Masculino , Examen Físico/métodos , Sensibilidad y Especificidad , Máquina de Vectores de SoporteRESUMEN
Segmentation and mutant classification of high-frequency ultrasound (HFU) mouse embryo brain ventricle (BV) and body images can provide valuable information for developmental biologists. However, manual segmentation and identification of BV and body requires substantial time and expertise. This article proposes an accurate, efficient and explainable deep learning pipeline for automatic segmentation and classification of the BV and body. For segmentation, a two-stage framework is implemented. The first stage produces a low-resolution segmentation map, which is then used to crop a region of interest (ROI) around the target object and serve as the probability map of the autocontext input for the second-stage fine-resolution refinement network. The segmentation then becomes tractable on high-resolution 3-D images without time-consuming sliding windows. The proposed segmentation method significantly reduces inference time (102.36-0.09 s/volume ≈ 1000× faster) while maintaining high accuracy comparable to previous sliding-window approaches. Based on the BV and body segmentation map, a volumetric convolutional neural network (CNN) is trained to perform a mutant classification task. Through backpropagating the gradients of the predictions to the input BV and body segmentation map, the trained classifier is found to largely focus on the region where the Engrailed-1 (En1) mutation phenotype is known to manifest itself. This suggests that gradient backpropagation of deep learning classifiers may provide a powerful tool for automatically detecting unknown phenotypes associated with a known genetic mutation.
Asunto(s)
Aprendizaje Profundo , Imagenología Tridimensional , Animales , Procesamiento de Imagen Asistido por Computador , Ratones , Redes Neurales de la Computación , UltrasonografíaRESUMEN
The segmentation of the brain ventricle (BV) and body in embryonic mice high-frequency ultrasound (HFU) volumes can provide useful information for biological researchers. However, manual segmentation of the BV and body requires substantial time and expertise. This work proposes a novel deep learning based end-to-end auto-context refinement framework, consisting of two stages. The first stage produces a low resolution segmentation of the BV and body simultaneously. The resulting probability map for each object (BV or body) is then used to crop a region of interest (ROI) around the target object in both the original image and the probability map to provide context to the refinement segmentation network. Joint training of the two stages provides significant improvement in Dice Similarity Coefficient (DSC) over using only the first stage (0.818 to 0.906 for the BV, and 0.919 to 0.934 for the body). The proposed method significantly reduces the inference time (102.36 to 0.09 s/volume ≈1000x faster) while slightly improves the segmentation accuracy over the previous methods using slide-window approaches.