Accurate estimation of biological age and its application in disease prediction using a multimodal image Transformer system.

Aging in an individual refers to the temporal change, mostly decline, in the body's ability to meet physiological demands. Biological age (BA) is a biomarker of chronological aging and can be used to stratify populations to predict certain age-related chronic diseases. BA can be predicted from biomedical features such as brain MRI, retinal, or facial images, but the inherent heterogeneity in the aging process limits the usefulness of BA predicted from individual body systems. In this paper, we developed a multimodal Transformer-based architecture with cross-attention which was able to combine facial, tongue, and retinal images to estimate BA. We trained our model using facial, tongue, and retinal images from 11,223 healthy subjects and demonstrated that using a fusion of the three image modalities achieved the most accurate BA predictions. We validated our approach on a test population of 2,840 individuals with six chronic diseases and obtained significant difference between chronological age and BA (AgeDiff) than that of healthy subjects. We showed that AgeDiff has the potential to be utilized as a standalone biomarker or conjunctively alongside other known factors for risk stratification and progression prediction of chronic diseases. Our results therefore highlight the feasibility of using multimodal images to estimate and interrogate the aging process.

CHARACTERIZATION BY FRACTAL DIMENSION ANALYSIS OF THE RETINAL CAPILLARY NETWORK IN PARKINSON DISEASE.

PURPOSE: To characterize retinal capillary complexity by optical coherence tomography angiography in Parkinson disease. METHOD: Twenty-five Parkinson disease patients and 25 age- and gender-matched healthy controls were recruited. Optical coherence tomography angiography and optical coherence tomography imaged the superficial and deep retinal capillary plexuses and retinal structure. Retinal capillary skeleton density, retinal capillary perfusion density, and fractal dimension analysis of retinal capillary complexity were performed in the total annular zone and quadrant sectors. The thickness of retinal nerve fiber layer, ganglion cell layer and inner plexiform layer, and total retinal thickness were extracted from retinal structural images. Relationships among the retinal capillaries, retinal structure, and clinical parameters were analyzed. RESULTS: The superficial retinal capillary plexus in Parkinson disease patients had lower retinal capillary skeleton and perfusion densities and capillary complexity in the total annular zone and all quadrant sectors compared with healthy control subjects. The deep retinal capillary plexus retinal capillary complexity was decreased in the total annular zone and the superior and inferior quadrants. The retinal capillary complexity in the inferior quadrant was negatively correlated with the best-corrected visual acuity and disease duration (r = -0.61, r = -0.43, respectively, both P < 0.05). CONCLUSION: As determined by fractal analysis, retinal capillary complexity can be an objective biomarker in Parkinson disease.

Artificial intelligence system for outcome evaluations of human in vitro fertilization-derived embryos.

BACKGROUND: In vitro fertilization (IVF) has emerged as a transformative solution for infertility. However, achieving favorable live-birth outcomes remains challenging. Current clinical IVF practices in IVF involve the collection of heterogeneous embryo data through diverse methods, including static images and temporal videos. However, traditional embryo selection methods, primarily reliant on visual inspection of morphology, exhibit variability and are contingent on the experience of practitioners. Therefore, an automated system that can evaluate heterogeneous embryo data to predict the final outcomes of live births is highly desirable. METHODS: We employed artificial intelligence (AI) for embryo morphological grading, blastocyst embryo selection, aneuploidy prediction, and final live-birth outcome prediction. We developed and validated the AI models using multitask learning for embryo morphological assessment, including pronucleus type on day 1 and the number of blastomeres, asymmetry, and fragmentation of blastomeres on day 3, using 19,201 embryo photographs from 8271 patients. A neural network was trained on embryo and clinical metadata to identify good-quality embryos for implantation on day 3 or day 5, and predict live-birth outcomes. Additionally, a 3D convolutional neural network was trained on 418 time-lapse videos of preimplantation genetic testing (PGT)-based ploidy outcomes for the prediction of aneuploidy and consequent live-birth outcomes. RESULTS: These two approaches enabled us to automatically assess the implantation potential. By combining embryo and maternal metrics in an ensemble AI model, we evaluated live-birth outcomes in a prospective cohort that achieved higher accuracy than experienced embryologists (46.1% vs. 30.7% on day 3, 55.0% vs. 40.7% on day 5). Our results demonstrate the potential for AI-based selection of embryos based on characteristics beyond the observational abilities of human clinicians (area under the curve: 0.769, 95% confidence interval: 0.709-0.820). These findings could potentially provide a noninvasive, high-throughput, and low-cost screening tool to facilitate embryo selection and achieve better outcomes. CONCLUSIONS: Our study underscores the AI model's ability to provide interpretable evidence for clinicians in assisted reproduction, highlighting its potential as a noninvasive, efficient, and cost-effective tool for improved embryo selection and enhanced IVF outcomes. The convergence of cutting-edge technology and reproductive medicine has opened new avenues for addressing infertility challenges and optimizing IVF success rates.

Accurate prediction of myopic progression and high myopia by machine learning.

Background: Myopia is a leading cause of visual impairment in Asia and worldwide. However, accurately predicting the progression of myopia and the high risk of myopia remains a challenge. This study aims to develop a predictive model for the development of myopia. Methods: We first retrospectively gathered 612 530 medical records from five independent cohorts, encompassing 227 543 patients ranging from infants to young adults. Subsequently, we developed a multivariate linear regression algorithm model to predict the progression of myopia and the risk of high myopia. Result: The model to predict the progression of myopia achieved an R2 value of 0.964 vs a mean absolute error (MAE) of 0.119D [95% confidence interval (CI): 0.119, 1.146] in the internal validation set. It demonstrated strong generalizability, maintaining consistent performance across external validation sets: R2 = 0.950 vs MAE = 0.119D (95% CI: 0.119, 1.136) in validation study 1, R2 = 0.950 vs MAE = 0.121D (95% CI: 0.121, 1.144) in validation study 2, and R2 = 0.806 vs MAE = -0.066D (95% CI: -0.066, 0.569) in the Shanghai Children Myopia Study. In the Beijing Children Eye Study, the model achieved an R2 of 0.749 vs a MAE of 0.178D (95% CI: 0.178, 1.557). The model to predict the risk of high myopia achieved an area under the curve (AUC) of 0.99 in the internal validation set and consistently high area under the curve values of 0.99, 0.99, 0.96 and 0.99 in the respective external validation sets. Conclusion: Our study demonstrates accurate prediction of myopia progression and risk of high myopia providing valuable insights for tailoring strategies to personalize and optimize the clinical management of myopia in children.

SARS-CoV-2 Delta and Omicron variants resist spike cleavage by human airway trypsin-like protease.

Soluble host factors in the upper respiratory tract can serve as the first line of defense against SARS-CoV-2 infection. In this study, we described the identification and function of a human airway trypsin-like protease (HAT), capable of reducing the infectivity of ancestral SARS-CoV-2. Further, in mouse models, HAT analogue expression was upregulated by SARS-CoV-2 infection. The antiviral activity of HAT functioned through the cleavage of the SARS-CoV-2 spike glycoprotein at R682. This cleavage resulted in inhibition of the attachment of ancestral spike proteins to host cells, which inhibited the cell-cell membrane fusion process. Importantly, exogenous addition of HAT notably reduced the infectivity of ancestral SARS-CoV-2 in vivo. However, HAT was ineffective against the Delta variant and most circulating Omicron variants, including the BQ.1.1 and XBB.1.5 subvariants. We demonstrate that the P681R mutation in Delta and P681H mutation in the Omicron variants, adjacent to the R682 cleavage site, contributed to HAT resistance. Our study reports what we believe to be a novel soluble defense factor against SARS-CoV-2 and resistance of its actions in the Delta and Omicron variants.

Multi-Learner Based Deep Meta-Learning for Few-Shot Medical Image Classification.

Few-shot learning (FSL) is promising in the field of medical image analysis due to high cost of establishing high-quality medical datasets. Many FSL approaches have been proposed in natural image scenes. However, present FSL methods are rarely evaluated on medical images and the FSL technology applicable to medical scenarios need to be further developed. Meta-learning has supplied an optional framework to address the challenging FSL setting. In this paper, we propose a novel multi-learner based FSL method for multiple medical image classification tasks, combining meta-learning with transfer-learning and metric-learning. Our designed model is composed of three learners, including auto-encoder, metric-learner and task-learner. In transfer-learning, all the learners are trained on the base classes. In the ensuing meta-learning, we leverage multiple novel tasks to fine-tune the metric-learner and task-learner in order to fast adapt to unseen tasks. Moreover, to further boost the learning efficiency of our model, we devised real-time data augmentation and dynamic Gaussian disturbance soft label (GDSL) scheme as effective generalization strategies of few-shot classification tasks. We have conducted experiments for three-class few-shot classification tasks on three newly-built challenging medical benchmarks, BLOOD, PATH and CHEST. Extensive comparisons to related works validated that our method achieved top performance both on homogeneous medical datasets and cross-domain datasets.

An Annotation-Free Restoration Network for Cataractous Fundus Images.

Cataracts are the leading cause of vision loss worldwide. Restoration algorithms are developed to improve the readability of cataract fundus images in order to increase the certainty in diagnosis and treatment for cataract patients. Unfortunately, the requirement of annotation limits the application of these algorithms in clinics. This paper proposes a network to annotation-freely restore cataractous fundus images (ArcNet) so as to boost the clinical practicability of restoration. Annotations are unnecessary in ArcNet, where the high-frequency component is extracted from fundus images to replace segmentation in the preservation of retinal structures. The restoration model is learned from the synthesized images and adapted to real cataract images. Extensive experiments are implemented to verify the performance and effectiveness of ArcNet. Favorable performance is achieved using ArcNet against state-of-the-art algorithms, and the diagnosis of ocular fundus diseases in cataract patients is promoted by ArcNet. The capability of properly restoring cataractous images in the absence of annotated data promises the proposed algorithm outstanding clinical practicability.

Dynamic analysis of iris changes and a deep learning system for automated angle-closure classification based on AS-OCT videos.

BACKGROUND: To study the association between dynamic iris change and primary angle-closure disease (PACD) with anterior segment optical coherence tomography (AS-OCT) videos and develop an automated deep learning system for angle-closure screening as well as validate its performance. METHODS: A total of 369 AS-OCT videos (19,940 frames)-159 angle-closure subjects and 210 normal controls (two datasets using different AS-OCT capturing devices)-were included. The correlation between iris changes (pupil constriction) and PACD was analyzed based on dynamic clinical parameters (pupil diameter) under the guidance of a senior ophthalmologist. A temporal network was then developed to learn discriminative temporal features from the videos. The datasets were randomly split into training, and test sets and fivefold stratified cross-validation were used to evaluate the performance. RESULTS: For dynamic clinical parameter evaluation, the mean velocity of pupil constriction (VPC) was significantly lower in angle-closure eyes (0.470 mm/s) than in normal eyes (0.571 mm/s) (P < 0.001), as was the acceleration of pupil constriction (APC, 3.512 mm/s2 vs. 5.256 mm/s2; P < 0.001). For our temporal network, the areas under the curve of the system using AS-OCT images, original AS-OCT videos, and aligned AS-OCT videos were 0.766 (95% CI: 0.610-0.923) vs. 0.820 (95% CI: 0.680-0.961) vs. 0.905 (95% CI: 0.802-1.000) (for Casia dataset) and 0.767 (95% CI: 0.620-0.914) vs. 0.837 (95% CI: 0.713-0.961) vs. 0.919 (95% CI: 0.831-1.000) (for Zeiss dataset). CONCLUSIONS: The results showed, comparatively, that the iris of angle-closure eyes stretches less in response to illumination than in normal eyes. Furthermore, the dynamic feature of iris motion could assist in angle-closure classification.

Risk Factors of Vitamin D Deficiency in Chinese Ischemic Stroke Patients: A Cross-Sectional Study.

PURPOSE: Lower serum vitamin D has been reported to be associated with stroke. This study aimed to analyze the risk factors of vitamin deficiency in Chinese stroke patients, and further analyze its impact in different gender and their clinical variables. METHODS: 982 stroke patients were enrolled. Laboratory parameters such as serum vitamin D, apolipoprotein A-I (ApoA-I), apolipoprotein B (ApoB), ApoA-I/ApoB, cholesterol (CH), fibrinogen (FIB), blood glucose (Glu), high-density lipoprotein (HDL), low-density lipoprotein cholesterol (LDL-C), and triglyceride (TG) were collected and recorded. The severity of stroke was assessed by National Institute of Health Stroke Scale (NIHSS) score. Based on their serum vitamin D level, patients were divided into three groups: Vitamin D deficiency (<50 nmol/L), vitamin D insufficiency (≥50-75 nmol/L) and vitamin D sufficiency (≥75 nmol/L) and differences were compared among the three groups. Statistical analyses were done to assess the risk factors for serum vitamin D deficiency in our ischemic stroke patients. RESULTS: Gender, NIHSS, and FIB showed significant differences among the vitamin D groups (P < 0.001 ∼ P = 0.002). The female gender (OR = 2.422, P < 0.001), severity of stroke using NIHSS (OR = 1.055, P = 0.008) and FIB (OR = 1.256, P = 0.005) were risk factors of vitamin D deficiency in ischemic stroke patients. In subgroup analysis, NIHSS was significantly associated with vitamin D deficiency in the male group (OR = 1.087, P = 0.002) and higher FIB group (OR = 1.078, P = 0.001). CONCLUSIONS: The female gender, severity of stroke using NIHSS and FIB were risk factors for vitamin D deficiency in our incident stroke patients. NIHSS was more sensitive to vitamin D deficiency in male ischemic stroke patients. Besides, under higher FIB circumstance, the increasing NIHSS score was more related to the vitamin D deficiency. Levels of vitamin D in patients with ischemic stroke should be well monitored during the disease cascade.

Altered resting-state functional connectivity density in patients with neuromyelitis optica-spectrum disorders.

BACKGROUND: Neuromyelitis optica spectrum disorder (NMOSD) is an autoimmune, demyelinating disorder, accompanied by abnormal spontaneous activity of the brain and impairment of the retina and optic nerve. Functional connectivity density (FCD) map, a graph theory method, was applied to explore the functional connectivity alterations of brian in NMOSD patients and investigate the alterations of FCD to the structural and microvascular changes around the optic nerve head (ONH). METHODS: Nineteen NMOSD patients and 22 healthy controls (HCs) were included in our study. All participants underwent resting-state functional magnetic resonance imaging (fMRI) scans of the brain, and ophthalmological examinations included optical coherence tomographic angiography (OCT-A) imaging, visual acuity (VA), and intraocular pressure (IOP). The long- and short-range FCD was calculated by the fMRI graph theory method and two-sample t-tests were performed to compare the discrepancy of FCD between NMOSD and HCs. OCT-A imaging was used to obtain the structure (peripapillary retinal nerve fiber layer, pRNFL) and microvessels (radial peripapillary capillary, RPC) details around the ONH. The association between the long- and short-range FCD values with the structural and microvascular variation around the ONH were evaluated using Spearman's correlation. RESULTS: Significantly decreased (corrected p < 0.05) long-range FCD was seen in the right superior parietal gyrus (SPG) in patients with NMOSD when compared to HCs. Increased long-range FCD was seen in the right fusiform gyrus (FFG), left orbital part of superior frontal orbital gyrus (ORBsup) and left anterior cingulum and paracingulate gyri (ACG) in NMOSD patients (corrected p < 0.05). The regions with reduced short-range FCD in NMOSD were the left angular gyrus (ANG) and right SPG (corrected p < 0.05). Increased short-range FCD was shown (corrected p < 0.05) in the right FFG of NMOSD. The pRNFL thickness and RPC density in all participants were negatively correlated with the long-range FCD values in the right FFG, left ORBsup, and left ACG as well as short-range FCD values in the right FFG, besides, both were positively correlated with the long-range FCD values in the right SPG and short-range FCD values in the left ANG and right SPG (p < 0.05). CONCLUSION: Our study demonstrates that patients with NMOSD have widespread brain dysfunction after optic neuritis attacks which shows as impairment of widespread spatial distribution in long- and short-range FCD. Structural and microvascular changes around the ONH are associated with neural changes in the brain.

Structural and Microvascular Changes in the Macular Are Associated With Severity of White Matter Lesions.

Purpose: This study aimed to characterize the microvascular and structural changes in the macular that occur in white matter hyperintensities (WMH) using optical coherence tomographic angiography. We also aimed to explore the association between macular microvascular and structural changes with focal markers of brain tissue on MRI in WMH using the Fazekas scale. Methods: This study enrolled healthy participants who were stroke- and dementia-free. MRI was used to image the cerebral white matter lesions, and Fazekas scale was used to evaluate the severity of the white matter lesions. Optical coherence tomography angiography (OCT-A) was used to image the radial peripapillary capillaries (RPCs), macular capillary plexuses [superficial capillary plexus (SCP) and deep capillary plexus (DCP)] and thickness around the optic nerve head, peripapillary retinal nerve fiber layer (pRNFL). Results: Seventy-four participants were enrolled and divided into two groups according to their Fazekas score (Fazekas scores ≤ 1 and ≥2). Participants with Fazekas score ≥2 showed significantly reduced RPC density (P = 0.02) and DCP density (P = 0.012) when compared with participants with Fazekas score ≤ 1. Participants with Fazekas score ≥2 showed reduced pRNFL (P = 0.004) when compared to participants with Fazekas score ≤ 1. Fazekas scores were significantly associated with the pRNFL thickness (Rho = -0.389, P = 0.001), RPC density (Rho = -0.248, P = 0.035), and DCP density (Rho = -0.283, P = 0.015), respectively. Conclusions: Microvascular impairment and neuro-axonal damage are associated with the disease cascade in WMH. We have shown that RPC and DCP densities are significantly affected, and these impairments are associated with the severity of the disease and cognitive function. OCT-A could be a useful tool in quantifying the retinal capillary densities in WMH.

Alterations in the Brain Structure and Functional Connectivity in Aquaporin-4 Antibody-Positive Neuromyelitis Optica Spectrum Disorder.

PURPOSE: To investigate the mechanisms underlying the gray matter volume (GMV) and functional connectivity (FC) changes in aquaporin-4 antibody-positive neuromyelitis optica spectrum disorder (NMOSD) patients. METHODS: This cross-sectional study consisted of 21 patients with aquaporin-4 antibody-positive NMOSD and 22 age- and sex-matched healthy controls. All participants underwent cerebral magnetic resonance imaging and testing each individual's visual acuity was done. RESULTS: Neuromyelitis optica spectrum disorder patients showed significantly reduced GMV in the left calcarine, left thalamus and right lingual gyrus of the NMOSD patients when compared to HC (P < 0.05). NMOSD patients showed significantly decreased FC values (P < 0.05) in both the left and right calcarine, right lingual gyrus and left thalamus, respectively, when compared to HC. We also observed a positive correlation between the FC values of the left thalamus, bilateral calcarine gyrus and the visual acuity, respectively (P < 0.05). Furthermore, a negative association was seen between the duration of the disease, frequency of optic neuritis, and the FC values in the lingual gyrus, bilateral calcarine gyrus, and right lingual gyrus, respectively (P < 0.05). CONCLUSION: Reduced visual acuity and frequency of optic neuritis are associated with alterations in the GMV and FC in NMOSD. Our current study, which provides imaging evidence on the impairment involved in NMOSD, sheds light on pathophysiological responses of optic neuritis attack on the brain especially on the visual network.