Screening of Moyamoya Disease From Retinal Photographs: Development and Validation of Deep Learning Algorithms.

BACKGROUND: Moyamoya disease (MMD) is a rare and complex pathological condition characterized by an abnormal collateral circulation network in the basal brain. The diagnosis of MMD and its progression is unpredictable and influenced by many factors. MMD can affect the blood vessels supplying the eyes, resulting in a range of ocular symptoms. In this study, we developed a deep learning model using real-world data to assist a diagnosis and determine the stage of the disease using retinal photographs. METHODS: This retrospective observational study conducted from August 2006 to March 2022 included 498 retinal photographs from 78 patients with MMD and 3835 photographs from 1649 healthy participants. Photographs were preprocessed, and an ResNeXt50 model was developed. Model performance was measured using receiver operating curves and their area under the receiver operating characteristic curve, accuracy, sensitivity, and F1-score. Heatmaps and progressive erasing plus progressive restoration were performed to validate the faithfulness. RESULTS: Overall, 322 retinal photographs from 67 patients with MMD and 3752 retinal photographs from 1616 healthy participants were used to develop a screening and stage prediction model for MMD. The average age of the patients with MMD was 44.1 years, and the average follow-up time was 115 months. Stage 3 photographs were the most prevalent, followed by stages 4, 5, 2, 1, and 6 and healthy. The MMD screening model had an average area under the receiver operating characteristic curve of 94.6%, with 89.8% sensitivity and 90.4% specificity at the best cutoff point. MMD stage prediction models had an area under the receiver operating characteristic curve of 78% or higher, with stage 3 performing the best at 93.6%. Heatmap identified the vascular region of the fundus as important for prediction, and progressive erasing plus progressive restoration result shows an area under the receiver operating characteristic curve of 70% only with 50% of the important regions. CONCLUSIONS: This study demonstrated that retinal photographs could be used as potential biomarkers for screening and staging of MMD and the disease stage could be classified by a deep learning algorithm.

Comprehensive Assessment and Early Prediction of Gross Motor Performance in Toddlers With Graph Convolutional Networks-Based Deep Learning: Development and Validation Study.

BACKGROUND: Accurate and timely assessment of children's developmental status is crucial for early diagnosis and intervention. More accurate and automated developmental assessments are essential due to the lack of trained health care providers and imprecise parental reporting. In various areas of development, gross motor development in toddlers is known to be predictive of subsequent childhood developments. OBJECTIVE: The purpose of this study was to develop a model to assess gross motor behavior and integrate the results to determine the overall gross motor status of toddlers. This study also aimed to identify behaviors that are important in the assessment of overall gross motor skills and detect critical moments and important body parts for the assessment of each behavior. METHODS: We used behavioral videos of toddlers aged 18-35 months. To assess gross motor development, we selected 4 behaviors (climb up the stairs, go down the stairs, throw the ball, and stand on 1 foot) that have been validated with the Korean Developmental Screening Test for Infants and Children. In the child behavior videos, we estimated each child's position as a bounding box and extracted human keypoints within the box. In the first stage, the videos with the extracted human keypoints of each behavior were evaluated separately using a graph convolutional networks (GCN)-based algorithm. The probability values obtained for each label in the first-stage model were used as input for the second-stage model, the extreme gradient boosting (XGBoost) algorithm, to predict the overall gross motor status. For interpretability, we used gradient-weighted class activation mapping (Grad-CAM) to identify important moments and relevant body parts during the movements. The Shapley additive explanations method was used for the assessment of variable importance, to determine the movements that contributed the most to the overall developmental assessment. RESULTS: Behavioral videos of 4 gross motor skills were collected from 147 children, resulting in a total of 2395 videos. The stage-1 GCN model to evaluate each behavior had an area under the receiver operating characteristic curve (AUROC) of 0.79 to 0.90. Keypoint-mapping Grad-CAM visualization identified important moments in each behavior and differences in important body parts. The stage-2 XGBoost model to assess the overall gross motor status had an AUROC of 0.90. Among the 4 behaviors, "go down the stairs" contributed the most to the overall developmental assessment. CONCLUSIONS: Using movement videos of toddlers aged 18-35 months, we developed objective and automated models to evaluate each behavior and assess each child's overall gross motor performance. We identified the important behaviors for assessing gross motor performance and developed methods to recognize important moments and body parts while evaluating gross motor performance.

Development and Validation of a Joint Attention-Based Deep Learning System for Detection and Symptom Severity Assessment of Autism Spectrum Disorder.

Importance: Joint attention, composed of complex behaviors, is an early-emerging social function that is deficient in children with autism spectrum disorder (ASD). Currently, no methods are available for objectively quantifying joint attention. Objective: To train deep learning (DL) models to distinguish ASD from typical development (TD) and to differentiate ASD symptom severities using video data of joint attention behaviors. Design, Setting, and Participants: In this diagnostic study, joint attention tasks were administered to children with and without ASD, and video data were collected from multiple institutions from August 5, 2021, to July 18, 2022. Of 110 children, 95 (86.4%) completed study measures. Enrollment criteria were 24 to 72 months of age and ability to sit with no history of visual or auditory deficits. Exposures: Children were screened using the Childhood Autism Rating Scale. Forty-five children were diagnosed with ASD. Three types of joint attention were assessed using a specific protocol. Main Outcomes and Measures: Correctly distinguishing ASD from TD and different levels of ASD symptom severity using the DL model area under the receiver operating characteristic curve (AUROC), accuracy, precision, and recall. Results: The analytical population consisted of 45 children with ASD (mean [SD] age, 48.0 [13.4] months; 24 [53.3%] boys) vs 50 with TD (mean [SD] age, 47.9 [12.5] months; 27 [54.0%] boys). The DL ASD vs TD models showed good predictive performance for initiation of joint attention (IJA) (AUROC, 99.6% [95% CI, 99.4%-99.7%]; accuracy, 97.6% [95% CI, 97.1%-98.1%]; precision, 95.5% [95% CI, 94.4%-96.5%]; and recall, 99.2% [95% CI, 98.7%-99.6%]), low-level response to joint attention (RJA) (AUROC, 99.8% [95% CI, 99.6%-99.9%]; accuracy, 98.8% [95% CI, 98.4%-99.2%]; precision, 98.9% [95% CI, 98.3%-99.4%]; and recall, 99.1% [95% CI, 98.6%-99.5%]), and high-level RJA (AUROC, 99.5% [95% CI, 99.2%-99.8%]; accuracy, 98.4% [95% CI, 97.9%-98.9%]; precision, 98.8% [95% CI, 98.2%-99.4%]; and recall, 98.6% [95% CI, 97.9%-99.2%]). The DL-based ASD symptom severity models showed reasonable predictive performance for IJA (AUROC, 90.3% [95% CI, 88.8%-91.8%]; accuracy, 84.8% [95% CI, 82.3%-87.2%]; precision, 76.2% [95% CI, 72.9%-79.6%]; and recall, 84.8% [95% CI, 82.3%-87.2%]), low-level RJA (AUROC, 84.4% [95% CI, 82.0%-86.7%]; accuracy, 78.4% [95% CI, 75.0%-81.7%]; precision, 74.7% [95% CI, 70.4%-78.8%]; and recall, 78.4% [95% CI, 75.0%-81.7%]), and high-level RJA (AUROC, 84.2% [95% CI, 81.8%-86.6%]; accuracy, 81.0% [95% CI, 77.3%-84.4%]; precision, 68.6% [95% CI, 63.8%-73.6%]; and recall, 81.0% [95% CI, 77.3%-84.4%]). Conclusions and Relevance: In this diagnostic study, DL models for identifying ASD and differentiating levels of ASD symptom severity were developed and the premises for DL-based predictions were visualized. The findings suggest that this method may allow digital measurement of joint attention; however, follow-up studies are necessary for further validation.

Three-dimensional label-free morphology of CD8 + T cells as a sepsis biomarker.

Sepsis is a dysregulated immune response to infection that leads to organ dysfunction and is associated with a high incidence and mortality rate. The lack of reliable biomarkers for diagnosing and prognosis of sepsis is a major challenge in its management. We aimed to investigate the potential of three-dimensional label-free CD8 + T cell morphology as a biomarker for sepsis. This study included three-time points in the sepsis recovery cohort (N = 8) and healthy controls (N = 20). Morphological features and spatial distribution within cells were compared among the patients' statuses. We developed a deep learning model to predict the diagnosis and prognosis of sepsis using the internal cell morphology. Correlation between the morphological features and clinical indices were analysed. Cell morphological features and spatial distribution differed significantly between patients with sepsis and healthy controls and between the survival and non-survival groups. The model for predicting the diagnosis and prognosis of sepsis showed an area under the receiver operating characteristic curve of nearly 100% with only a few cells, and a strong correlation between the morphological features and clinical indices was observed. Our study highlights the potential of three-dimensional label-free CD8 + T cell morphology as a promising biomarker for sepsis. This approach is rapid, requires a minimum amount of blood samples, and has the potential to provide valuable information for the early diagnosis and prognosis of sepsis.

Development of Deep Ensembles to Screen for Autism and Symptom Severity Using Retinal Photographs.

Importance: Screening for autism spectrum disorder (ASD) is constrained by limited resources, particularly trained professionals to conduct evaluations. Individuals with ASD have structural retinal changes that potentially reflect brain alterations, including visual pathway abnormalities through embryonic and anatomic connections. Whether deep learning algorithms can aid in objective screening for ASD and symptom severity using retinal photographs is unknown. Objective: To develop deep ensemble models to differentiate between retinal photographs of individuals with ASD vs typical development (TD) and between individuals with severe ASD vs mild to moderate ASD. Design, Setting, and Participants: This diagnostic study was conducted at a single tertiary-care hospital (Severance Hospital, Yonsei University College of Medicine) in Seoul, Republic of Korea. Retinal photographs of individuals with ASD were prospectively collected between April and October 2022, and those of age- and sex-matched individuals with TD were retrospectively collected between December 2007 and February 2023. Deep ensembles of 5 models were built with 10-fold cross-validation using the pretrained ResNeXt-50 (32×4d) network. Score-weighted visual explanations for convolutional neural networks, with a progressive erasing technique, were used for model visualization and quantitative validation. Data analysis was performed between December 2022 and October 2023. Exposures: Autism Diagnostic Observation Schedule-Second Edition calibrated severity scores (cutoff of 8) and Social Responsiveness Scale-Second Edition T scores (cutoff of 76) were used to assess symptom severity. Main Outcomes and Measures: The main outcomes were participant-level area under the receiver operating characteristic curve (AUROC), sensitivity, and specificity. The 95% CI was estimated through the bootstrapping method with 1000 resamples. Results: This study included 1890 eyes of 958 participants. The ASD and TD groups each included 479 participants (945 eyes), had a mean (SD) age of 7.8 (3.2) years, and comprised mostly boys (392 [81.8%]). For ASD screening, the models had a mean AUROC, sensitivity, and specificity of 1.00 (95% CI, 1.00-1.00) on the test set. These models retained a mean AUROC of 1.00 using only 10% of the image containing the optic disc. For symptom severity screening, the models had a mean AUROC of 0.74 (95% CI, 0.67-0.80), sensitivity of 0.58 (95% CI, 0.49-0.66), and specificity of 0.74 (95% CI, 0.67-0.82) on the test set. Conclusions and Relevance: These findings suggest that retinal photographs may be a viable objective screening tool for ASD and possibly for symptom severity. Retinal photograph use may speed the ASD screening process, which may help improve accessibility to specialized child psychiatry assessments currently strained by limited resources.