Jane is a 7-year-old girl who presented for a developmental-behavioral pediatrics (DBP) evaluation because of parental concern about developmental delays and atypical behavior. Jane is a former term infant who was small for gestational age. She experienced some initial feeding difficulties and was discharged home from the nursery at 5 days of age. As a toddler, Jane was mildly delayed in acquiring gross motor developmental milestones, but she was more significantly delayed in acquiring fine motor and adaptive developmental milestones. She began receiving private occupational therapy (OT) services at 18 months of age. Jane's mother reported persistent current concerns about Jane's fine motor and adaptive development, reporting that she is unable to unbutton, button, or tie her shoes. She reported that at school, Jane has difficulty with handwriting and drawing, and she has had a particular academic difficulty in math. She also reported that despite not exhibiting delays in her acquisition of speech and language developmental milestones, having a large vocabulary, and being able to relate experiences verbally, Jane has trouble with back-and-forth conversation. She reported that Jane frequently makes tangential comments, and she tends to have one-sided conversations, where she will repetitively list facts about her restricted interest in insects, without acknowledging whether others are listening or paying attention to her. She reported that Jane also tends to take things very literally, and she has difficulty recognizing humor and sarcasm. Jane's mother reported that Jane exhibits behavioral rigidity, with a strict need for routine and an overfocus on rules, and she often tattles on other children for breaking the rules. She reported that Jane prefers to play by herself, and she does not have any friends at school or in her neighborhood. She reported that Jane has no difficulties with inattention, impulsivity, or hyperactivity at home or at school. On physical examination, Jane was found to have short stature, webbing of her neck, and a low posterior hairline.During the DBP evaluation, Jane was observed to exhibit poor eye contact, to have difficulty modulating the volume of her voice, and to engage in one-sided conversations. On developmental testing, Jane received a standard score of 72 (age equivalent of 4 years, 6 months) in visual-motor integration on the Beery-Buktenica Developmental Test of Visual Motor Integration. On the Kaufman Brief Intelligence Test-2, she received a verbal standard score of 121 and a nonverbal standard score of 88. On the Wide Range Achievement Test-5, she received a Word Reading standard score of 129, a Spelling standard score of 126, and a Math Computation standard score of 74. Based on the history provided by her mother and direct observations of her behavior during the DBP evaluation, on the Childhood Autism Rating Scale, Second Edition, High Functioning Version, Jane received a score of 34, placing her behavior in the severe symptoms of autism spectrum disorder category. Given her physical examination and the results of her developmental evaluation, genetic testing was pursued, and Jane was found to have Turner syndrome (monosomy X).Based on her diagnosis of Turner syndrome, Jane was determined by her school district to qualify for an Individualized Education Program under the category of "Other Health Impairment," and she was provided direct OT services and accommodations for math. She was not qualified to receive special education services under either the "Autism" or "Specific Learning Disability" categories, as it was concluded by her special education team that her medical diagnosis of Turner syndrome accounted for all of her special educational needs.What diagnoses need to be included in the DBP report, and what education does the medical team need to provide the special education team regarding the relationship between etiological and developmental diagnoses for children with developmental disabilities?

Limbal Stem Cell Deficiency (LSCD) is a corneal disorder characterized by conjunctivalization of the cornea. Central corneal epithelial thickness, measured using Anterior Segment Optical Coherence Tomography (AS-OCT), is a biomarker for LSCD. Manual measurement of epithelial thickness remains the gold standard for diagnosis because existing automated commercial software is unreliable in severe cases. Standard machine learning approaches have limited applicability for automating this measurement due to the difficulty of distinguishing corneal epithelium from scar tissue, and imprecise clinical labels. The goal of this study is to develop a deep learning pipeline to automate measurement of central epithelial thickness in AS-OCT images of eyes with LSCD. To accomplish this goal, we developed AI-CE, a model based on Multi-scale U-Net (MSU-Net) trained to accurately delineate the epithelial layer. The primary dataset comprised of 134 patient eyes, including 47 healthy normal eyes (35.0%), 34 with Stage 1 (25.4%), 34 with Stage 2 (25.4%), and 19 with Stage 3 (14.2%) LSCD. Performance was evaluated by calculating the average mean absolute error between model segmentations and manually measured clinical assessments. AI-CE model performance was compared against a commercially available segmentation software to demonstrate its improvement over the current state-of-the-art. The proposed model achieved a mean absolute error of 5.10 μm, significantly improving (p < 0.01) over clinical software (12.20 μm). This result was verified in a secondary retrospective dataset which similarly showed significant improvement over commercial software. We also demonstrate that clinical epithelial thickness values provided by AI-CE can distinguish between LSCD stages similarly to manual evaluations.

Retinal diseases are a leading cause of visual impairment globally. Early detection is crucial for effective management, and recent advances in ocular imaging and deep learning have facilitated the development of automated diagnostic systems. Traditional deep learning methods often struggle to effectively capture the overall structural context of the retina and the subtle, localised pathological features required for accurate multi label classification. Moreover, their decision-making processes typically lack transparency, limiting clinical trust and adoption. In this study, we propose an explainable transformer-based fine-tuned model for retinal disease classification that uses a Vision Transformer architecture enhanced with a Multiple Instance Learning (MIL) mechanism. The model is fine-tuned on the multi label retinal disease (MuReD) dataset, enabling it to simultaneously predict multiple co-existing pathologies. Experimental evaluation on the MuReD dataset demonstrates state-of-the-art performance, with the proposed fine-tuned MIL-VT achieving an ML_F1 (multi label F1) score of 0.672, representing an approximately 10 percentage-point increase over recent transformer-based baselines, alongside higher ML_mAP (multi label mAP) and comparable AUC scores. Additionally, we evaluate our model's performance on an independent Eye Disease Image (EDI) dataset, where it maintains better accuracy, achieving a macro-average F1 of approximately 0.92, highlighting its robust generalisation across datasets. An ablation study further validates the effectiveness of the MIL head and selective fine-tuning strategy, showing their substantial contributions to performance gains. To enhance clinical interpretability, we incorporate Grad CAM visualisations that highlight pathological regions associated with each predicted condition. The combination of better classification performance, explainability, and cross-dataset robustness underscores the potential of our proposed framework to improve automated retinal disease screening in diverse clinical settings.

Ocular drug delivery systems characterized by extended retention time and high bioavailability are urgently required to overcome the limitations of conventional eye drops. In this study, a novel drug-loaded silicone hydrogel contact lens (SHCL) was developed utilizing hydroxyl-terminated polydimethylsiloxane (PDMS-OH) as the silicone matrix, poly(ethylene glycol)-grafted methacryloyldopamine (PEG-MAD) as a functional monomer, and sodium acetyl hyaluronate (SAH) as a surface coating. PEG-MAD was synthesized via a lipase-catalyzed reaction between carboxyl-terminated PEG (PEG-COOH) and methacryloyldopamine (MAD). This synthetic approach not only prevented the oxidation and yellowing of MAD during high-temperature polymerization but also introduced additional hydrogen bonds, thereby enhancing the mechanical properties and drug-loading capacity of the hydrogel. The SHCL was fabricated through free radical polymerization, followed by drug loading (using levofloxacin as a model drug) and SAH coating. The optimized SHCL exhibited a water content of 59.5%, high optical transmittance (91% in the visible light region), and no ocular irritation (as evaluated by the CAM test, with an irritation score of 0). Compared to control groups lacking PEG-MAD or using 3-(Methacryloyloxy)propyltris(trimethylsiloxy)silane (TRIS) as the silicone matrix, the developed SHCL demonstrated a higher drug-loading capacity (3.28 μg/mg dry gel) and sustained drug release over 96 h, along with a significant reduction in human serum albumin adsorption. These findings indicate that the PEG-MAD-functionalized, PDMS-OH-based SHCL holds considerable promise for clinical applications in ocular drug delivery.

Accurate registration between lateral cephalometric radiographs and facial profile photographs is a prerequisite for integrated craniofacial analysis, yet it remains challenging due to inherent modality gaps and the absence of clinically validated evaluation criteria. This study presents a clinically-driven engineering framework that systematically integrates high-resolution keypoint localization with a weighted Procrustes-based two-dimensional similarity registration. The uniform scale factor accommodates global magnification differences between imaging modalities, while shearing and local non-rigid deformation are excluded. To ensure clinical interpretability, we incorporate expert-defined anatomical priors as weights into the transformation process and introduce an adaptive seed-driven contour tracking (ASDCT) algorithm for precise performance evaluation. Experiments on 198 paired clinical samples demonstrate that our framework achieves stable performance (Contour Distance of 3.52 pixels; Contour Dilation Overlap of 87%) and provides metrics that correlate strongly with expert clinical assessments. By unifying landmark-guided alignment and boundary-consistency evaluation, this framework offers a robust, reproducible, and anatomically reliable solution for cross-modal craniofacial integration, effectively bridging the gap between computational precision and clinical practice.

Gestures can convey communicative and socially relevant information, but it remains unclear whether their spatial direction relative to the body modulates visual attention. Previous findings suggest that gestures directed towards the body (TB) may be particularly relevant for socio-cognitive processing. The present study investigated whether gesture direction and meaning modulate visual attention to the face and hand during gesture observation and imitation. Sixty video clips depicted intransitive gestures, pantomimes, and meaningless gestures directed either TB or away from the body (AB). Twenty-five healthy participants first observed and subsequently imitated the gestures while their eye movements were recorded. Average fixation duration within the face and hand areas of interest was analysed using linear mixed models. Fixation duration was longer during observation than imitation and was influenced by gesture type, with intransitive gestures eliciting longer fixations than pantomimes and meaningless gestures. Overall, the face attracted longer fixations than the hand, particularly during meaningful gestures. Although fixation durations were descriptively longer for TB than AB gestures, the effect of direction did not reach statistical significance. These findings suggest that visual attention during gesture processing is primarily shaped by gesture meaning and task demands, while the contribution of gesture direction to overt visual attention remains to be established.