Predicting mental and psychomotor delay in very pre-term infants using machine learning.

BACKGROUND: Very preterm infants are at elevated risk for neurodevelopmental delays. Earlier prediction of delays allows timelier intervention and improved outcomes. Machine learning (ML) was used to predict mental and psychomotor delay at 25 months. METHODS: We applied RandomForest classifier to data from 1109 very preterm infants recruited over 20 years. ML selected key predictors from 52 perinatal and 16 longitudinal variables (1-22 mo assessments). SHapley Additive exPlanations provided model interpretability. RESULTS: Balanced accuracy with perinatal variables was 62%/61% (mental/psychomotor). Top predictors of mental and psychomotor delay overlapped and included: birth year, days in hospital, antenatal MgSO4, days intubated, birth weight, abnormal cranial ultrasound, gestational age, mom's age and education, and intrauterine growth restriction. Highest balanced accuracy was achieved with 19-month follow-up scores and perinatal variables (72%/73%). CONCLUSIONS: Combining perinatal and longitudinal data, ML modeling predicted 24 month mental/psychomotor delay in very preterm infants ½ year early, allowing intervention to start that much sooner. Modeling using only perinatal features fell short of clinical application. Birth year's importance reflected a linear decline in predicting delay as birth year became more recent. IMPACT: Combining perinatal and longitudinal data, ML modeling was able to predict 24 month mental/psychomotor delay in very preterm infants ½ year early (25% of their lives) potentially advancing implementation of intervention services. Although cognitive/verbal and fine/gross motor delays require separate interventions, in very preterm infants there is substantial overlap in the risk factors that can be used to predict these delays. Birth year has an important effect on ML prediction of delay in very preterm infants, with those born more recently (1989-2009) being increasing less likely to be delayed, perhaps reflecting advances in medical practice.

Tafenoquine and its derivatives as inhibitors for the severe acute respiratory syndrome coronavirus 2.

The pandemic caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has severely affected human lives around the world as well as the global economy. Therefore, effective treatments against COVID-19 are urgently needed. Here, we screened a library containing Food and Drug Administration (FDA)-approved compounds to identify drugs that could target the SARS-CoV-2 main protease (Mpro), which is indispensable for viral protein maturation and regard as an important therapeutic target. We identified antimalarial drug tafenoquine (TFQ), which is approved for radical cure of Plasmodium vivax and malaria prophylaxis, as a top candidate to inhibit Mpro protease activity. The crystal structure of SARS-CoV-2 Mpro in complex with TFQ revealed that TFQ noncovalently bound to and reshaped the substrate-binding pocket of Mpro by altering the loop region (residues 139-144) near the catalytic Cys145, which could block the catalysis of its peptide substrates. We also found that TFQ inhibited human transmembrane protease serine 2 (TMPRSS2). Furthermore, one TFQ derivative, compound 7, showed a better therapeutic index than TFQ on TMPRSS2 and may therefore inhibit the infectibility of SARS-CoV-2, including that of several mutant variants. These results suggest new potential strategies to block infection of SARS-CoV-2 and rising variants.

Crystal structure and functional implication of a bacterial cyclic AMP-AMP-GMP synthetase.

Mammalian cyclic GMP-AMP synthase (cGAS) and its homologue dinucleotide cyclase in Vibrio cholerae (VcDncV) produce cyclic dinucleotides (CDNs) that participate in the defense against viral infection. Recently, scores of new cGAS/DncV-like nucleotidyltransferases (CD-NTases) were discovered, which produce various CDNs and cyclic trinucleotides (CTNs) as second messengers. Here, we present the crystal structures of EcCdnD, a CD-NTase from Enterobacter cloacae that produces cyclic AMP-AMP-GMP, in its apo-form and in complex with ATP, ADP and AMPcPP, an ATP analogue. Despite the similar overall architecture, the protein shows significant structural variations from other CD-NTases. Adjacent to the donor substrate, another nucleotide is bound to the acceptor binding site by a non-productive mode. Isothermal titration calorimetry results also suggest the presence of two ATP binding sites. GTP alone does not bind to EcCdnD, which however binds to pppApG, a possible intermediate. The enzyme is active on ATP or a mixture of ATP and GTP, and the best metal cofactor is Mg2+. The conserved residues Asp69 and Asp71 are essential for catalysis, as indicated by the loss of activity in the mutants. Based on structural analysis and comparison with VcDncV and RNA polymerase, a tentative catalytic pathway for the CTN-producing EcCdnD is proposed.

Real-world data on the prescription of proprotein convertase subtilisin/kexin type 9 inhibitors in high-risk patients in a tertiary medical center.

Proprotein convertase subtilisin/kexin type 9 (PCSK9) inhibitors, alirocumab and evolocumab, are currently approved for clinical use by Taiwan National Health Insurance (NHI) in patients who had a recent atherosclerotic cardiovascular disease with persistent LDL-C levels >135 mg/dL despite high-intensity statin (HIS) or maximally tolerated statin in combination with ezetimibe treatment. Since January 2020 to July 2020, total of 10 patients who had received coronary revascularization received NHI-approved alirocumab or evolocumab in our institution. The mean reduction of LDL-C following PCSK9 inhibitors treatment at 6-month and 12-month were respectively 62.5% and 60.2%. The patients in our study were younger, had more frequently received HIS/ezetimibe, and had higher baseline LDL-C levels with a greater LDL-C reduction following PCSK9 inhibitors treatment compared with those patients in previously studies. Our findings highlight that the NHI's regulation of PCSK9 inhibitors application should be re-evaluation to increase the use of NHI-approved PCSK9 inhibitors in high-risk patients.

Real-World Data on Comparison of Lipid-Lowering Effects between Alirocumab and Evolocumab in Patients with Hypercholesterolemia.

Background: Even though the proprotein convertase subtilisin/kexin type 9 (PCSK9) inhibitors, alirocumab and evolocumab, have been approved to reduce plasma low-density lipoprotein cholesterol (LDL-C) and the risk of atherosclerotic cardiovascular disease in high-risk patients, real-world data showing comparisons of the lipid-lowering effects between alirocumab and evolocumab are scarce because of the low prescription rates of PCSK9 inhibitors in clinical practice. Methods: Between Feb 2018 and Sep 2021, 22 patients who received alirocumab and 22 patients who received evolocumab at a tertiary medical center were enrolled. The patients' baseline characteristics, prescribed medications, plasma LDL-C levels, and percentages of reduction in LDL-C were compared between alirocumab users and evolocumab users. Results: The alirocumab users more frequently received ezetimibe treatment (72.7% vs. 40.9%, p = 0.03) and had higher baseline LDL-C (165.6 ± 63.2 mg/dL vs. 130.8 ± 56.3 mg/dL, p = 0.04) compared with the evolocumab users. At 6 months of follow-up, the plasma LDL-C levels in the alirocumab users were similar to those in the evolocumab users (79.3 ± 64.0 mg/dL vs. 63.5 ± 44.1 mg/dL, p = 0.48). Additionally, the percentages of LDL-C reduction following treatment were similar between the alirocumab users and evolocumab users (-47.0% ± 45.0% vs. -49.8 ± 24.9%, p = 0.66). Conclusions: The LDL-C lowering effects of alirocumab are similar to those of evolocumab in clinical practice.

Myopia Prevention and Outdoor Light Intensity in a School-Based Cluster Randomized Trial.

PURPOSE: To investigate the effectiveness of a school-based program promoting outdoor activities in Taiwan for myopia prevention and to identify protective light intensities. DESIGN: Multi-area, cluster-randomized intervention controlled trial. PARTICIPANTS: A total 693 grade 1 schoolchildren in 16 schools participated. Two hundred sixty-seven schoolchildren were in the intervention group and 426 were in the control group. METHODS: Initially, 24 schools were randomized into the intervention and control groups, but 5 and 3 schools in the intervention and control groups, respectively, withdrew before enrollment. A school-based Recess Outside Classroom Trial was implemented in the intervention group, in which schoolchildren were encouraged to go outdoors for up to 11 hours weekly. Data collection included eye examinations, cycloplegic refraction, noncontact axial length measurements, light meter recorders, diary logs, and questionnaires. MAIN OUTCOME MEASURES: Change in spherical equivalent and axial length after 1 year and the intensity and duration of outdoor light exposures. RESULTS: The intervention group showed significantly less myopic shift and axial elongation compared with the control group (0.35 diopter [D] vs. 0.47 D; 0.28 vs. 0.33 mm; P = 0.002 and P = 0.003) and a 54% lower risk of rapid myopia progression (odds ratio, 0.46; 95% confidence interval [CI], 0.28-0.77; P = 0.003). The myopic protective effects were significant in both nonmyopic and myopic children compared with controls. Regarding spending outdoor time of at least 11 hours weekly with exposure to 1000 lux or more of light, the intervention group had significantly more participants compared with the control group (49.79% vs. 22.73%; P < 0.001). Schoolchildren with longer outdoor time in school (≥200 minutes) showed significantly less myopic shift (measured by light meters; ≥1000 lux: 0.14 D; 95% CI, 0.02-0.27; P = 0.02; ≥3000 lux: 0.16 D; 95% CI, 0.002-0.32; P = 0.048). CONCLUSIONS: The school-based outdoor promotion program effectively reduced the myopia change in both nonmyopic and myopic children. Outdoor activities with strong sunlight exposure may not be necessary for myopia prevention. Relatively lower outdoor light intensity activity with longer time outdoors, such as in hallways or under trees, also can be considered.

High molecular weight of polysaccharides from Hericium erinaceus against amyloid beta-induced neurotoxicity.

BACKGROUND: Hericium erinaceus (HE) is a well-known mushroom in traditional Chinese food and medicine. HE extracts from the fruiting body and mycelia not only exhibit immunomodulatory, antimutagenic and antitumor activity but also have neuroprotective properties. Here, we purified HE polysaccharides (HEPS), composed of two high molecular weight polysaccharides (1.7 × 10(5) Da and 1.1 × 10(5) Da), and evaluated their protective effects on amyloid beta (Aß)-induced neurotoxicity in rat pheochromocytoma PC12 cells. METHODS: HEPS were prepared and purified using a 95 % ethanol extraction method. The components of HEPS were analyzed and the molecular weights of the polysaccharides were determined using high-pressure liquid chromatography (HPLC). The neuroprotective effects of the polysaccharides were evaluated through a 2,2-diphenyl-1-picrylhydrazyl (DPPH) radical scavenging assay and an MTT assay and by quantifying reactive oxygen species (ROS) and mitochondrial membrane potentials (MMP) of Aß-induced neurotoxicity in cells. RESULT: Our results showed that 250 µg/ml HEPS was harmless and promoted cell viability with 1.2 µM Aß treatment. We observed that the free radical scavenging rate exceeded 90 % when the concentration of HEPS was higher than 1 mg/mL in cells. The HEPS decreased the production of ROS from 80 to 58 % in a dose-dependent manner. Cell pretreatment with 250 µg/mL HEPS significantly reduced Aß-induced high MMPs from 74 to 51 % and 94 to 62 % at 24 and 48 h, respectively. Finally, 250 µg/mL of HEPS prevented Aß-induced cell shrinkage and nuclear degradation of PC12 cells. CONCLUSION: Our results demonstrate that HEPS exhibit antioxidant and neuroprotective effects on Aß-induced neurotoxicity in neurons.

Chondrogenesis in scleral stem/progenitor cells and its association with form-deprived myopia in mice.

PURPOSE: Previously, we demonstrated that scleral stem/progenitor cells (SSPCs) from mice have a chondrogenic differentiation potential, which is stimulated by transforming growth factor-ß (TGF-ß). In the present study, we hypothesized that chondrogenesis in the sclera could be a possible mechanism in myopia development. Therefore, we investigated the association of form-deprivation myopia (FDM) with expressions in mice sclera representing the chondrogenic phenotype: collagen type II (Col2) and α-smooth muscle actin (α-SMA). METHODS: The mRNA levels of α-SMA and Col2 in cultured murine SSPCs during chondrogenesis stimulated by TGF-ß2 were determined by real-time quantitative RT-PCR (qRT-PCR). The expression patterns of α-SMA and Col2 were assessed by immunohistochemistry in a three dimensional pellet culture. In an FDM mouse model, a western blot analysis and immunofluorescence study were used to detect the changes in the α-SMA and Col2 protein expressions in the sclera. In the RPE-choroid complex, qRT-PCR was used to detect any changes in the TGF-ß mRNA expression. RESULTS: The treatment of SSPCs in vitro with TGF-ß2 for 24 h at 1 or 10 ng/ml led to increased levels of both the α-SMA and Col2 expressions. In addition, we observed the formation of cartilage-like pellets from TGF-ß2-treated SSPCs. Both α-SMA and Col2 were expressed in the pellet. In an in-vivo study, the α-SMA and Col2 protein expressions were significantly increased in the sclera of FDM eyes in comparison to contralateral control eyes. Similarly, the levels of TGF-ß in the RPE-choroid complex of an FDM eye were also significantly elevated. CONCLUSION: Based on the concept of stem cells possessing multipotent differentiation potentials, scleral chondrogenesis induced by SSPCs may play a role in myopia development. The increased expressions of the cartilage-associated proteins Col2 and α-SMA during scleral chondrogenesis may be potential markers for myopia development. In addition, the increased levels of TGF-ß mRNA in the RPE-choroid complex might induce the chondrogenic change in the sclera during myopia development.

Sex-based differences in ischemic cardiovascular and bleeding outcomes following implantation of drug-eluting stent in patients at high bleeding risk.

BACKGROUND: Patients with high bleeding risk (HBR) may exhibit uncertain adherence to dual antiplatelet therapy (DAPT) following drug-eluting stent (DES) implantation. The current population-based cohort study aimed to investigate the sex-based differences in adverse outcomes among the HBR population by analyzing the National Health Insurance Research Database in Taiwan. METHODS: Patients who had HBR features defined by the Academic Research Consortium (ARC) and received DES implantation between January 1, 2007, and December 31, 2017, were enrolled. Propensity score matching was adopted to select 3,981 pairs with similar clinical cardiovascular risks but different sexes. A competing risk model was performed to evaluate the risk of adverse ischemic events (cardiac death, nonfatal myocardial infarction, and ischemic stroke) and any bleeding events in both sexes. Noncardiac death was considered a competing risk. RESULTS: Within a 5-year follow-up, the incidence rates (per 1,000 person-year (95% confidence interval (CI)) of composite ischemic events and any bleeding events in males were respectively 44.09 (40.25-48.30) and 42.55 (38.79-46.68), while those in females were respectively 40.18 (36.51-44.23) and 42.35 (38.57-46.51). After adjustment for clinical variables, male patients had a marginally increased risk in the composite ischemic events (adjusted subdistribution hazard ratio (SHR) = 1.15 (1.00-1.31), p = 0.045) and a similar risk of any bleeding events (adjusted SHR = 1.00 (0.88-1.15), p = 0.946) compared with female patients. CONCLUSIONS: Of the HBR population, males had an increased risk of ischemic outcomes but a similar risk of bleeding compared with females following DES implantation.

Magnetic resonance imaging-based deep learning imaging biomarker for predicting functional outcomes after acute ischemic stroke.

PURPOSE: Clinical risk scores are essential for predicting outcomes in stroke patients. The advancements in deep learning (DL) techniques provide opportunities to develop prediction applications using magnetic resonance (MR) images. We aimed to develop an MR-based DL imaging biomarker for predicting outcomes in acute ischemic stroke (AIS) and evaluate its additional benefit to current risk scores. METHOD: This study included 3338 AIS patients. We trained a DL model using deep neural network architectures on MR images and radiomics to predict poor functional outcomes at three months post-stroke. The DL model generated a DL score, which served as the DL imaging biomarker. We compared the predictive performance of this biomarker to five risk scores on a holdout test set. Additionally, we assessed whether incorporating the imaging biomarker into the risk scores improved the predictive performance. RESULTS: The DL imaging biomarker achieved an area under the receiver operating characteristic curve (AUC) of 0.788. The AUCs of the five studied risk scores were 0.789, 0.793, 0.804, 0.810, and 0.826, respectively. The imaging biomarker's predictive performance was comparable to four of the risk scores but inferior to one (p = 0.038). Adding the imaging biomarker to the risk scores improved the AUCs (p-values) to 0.831 (0.003), 0.825 (0.001), 0.834 (0.003), 0.836 (0.003), and 0.839 (0.177), respectively. The net reclassification improvement and integrated discrimination improvement indices also showed significant improvements (all p < 0.001). CONCLUSIONS: Using DL techniques to create an MR-based imaging biomarker is feasible and enhances the predictive ability of current risk scores.

Anomaly-guided weakly supervised lesion segmentation on retinal OCT images.

The availability of big data can transform the studies in biomedical research to generate greater scientific insights if expert labeling is available to facilitate supervised learning. However, data annotation can be labor-intensive and cost-prohibitive if pixel-level precision is required. Weakly supervised semantic segmentation (WSSS) with image-level labeling has emerged as a promising solution in medical imaging. However, most existing WSSS methods in the medical domain are designed for single-class segmentation per image, overlooking the complexities arising from the co-existence of multiple classes in a single image. Additionally, the multi-class WSSS methods from the natural image domain cannot produce comparable accuracy for medical images, given the challenge of substantial variation in lesion scales and occurrences. To address this issue, we propose a novel anomaly-guided mechanism (AGM) for multi-class segmentation in a single image on retinal optical coherence tomography (OCT) using only image-level labels. AGM leverages the anomaly detection and self-attention approach to integrate weak abnormal signals with global contextual information into the training process. Furthermore, we include an iterative refinement stage to guide the model to focus more on the potential lesions while suppressing less relevant regions. We validate the performance of our model with two public datasets and one challenging private dataset. Experimental results show that our approach achieves a new state-of-the-art performance in WSSS for lesion segmentation on OCT images.

Outdoor activity during class recess reduces myopia onset and progression in school children.

PURPOSE: The aim of this study was to investigate the effect of outdoor activity during class recess on myopia changes among elementary school students in a suburban area of Taiwan. DESIGN: Prospective, comparative, consecutive, interventional study. PARTICIPANTS: Elementary school students 7 to 11 years of age recruited from 2 nearby schools located in a suburban area of southern Taiwan. INTERVENTION: The children of one school participated in the interventions, whereas those from the other school served as the control group. The interventions consisted of performing a recess outside the classroom (ROC) program that encouraged children to go outside for outdoor activities during recess. The control school did not have any special programs during recess. MAIN OUTCOME MEASURES: Data were obtained by means of a parent questionnaire and ocular evaluations that included axial length and cycloplegic autorefraction at the beginning and after 1 year. RESULTS: Five hundred seventy-one students were recruited for this study, of whom 333 students participated in the interventional program, and 238 students were in the control school. At the beginning of the study, there were no significant differences between these 2 schools with regard to age, gender, baseline refraction, and myopia prevalence (47.75% vs. 49.16%). After 1 year, new onset of myopia was significantly lower in the ROC group than in the control group (8.41% vs. 17.65%; P<0.001). There was also significantly lower myopic shift in the ROC group compared with the control group (-0.25 diopter [D]/year vs. -0.38 D/year; P = 0.029). The multivariate analysis demonstrated that the variables of intervention of the ROC program and higher school year proved to be a protective factor against myopia shift in nonmyopic subjects (P = 0.020 and P = 0.017, respectively). For myopic subjects, school year was the only variable significantly associated with myopia progression (P = 0.006). CONCLUSIONS: Outdoor activities during class recess in school have a significant effect on myopia onset and myopic shift. Such activities have a prominent effect on the control of myopia shift, especially in nonmyopic children. FINANCIAL DISCLOSURE(S): The author(s) have no proprietary or commercial interest in any materials discussed in this article.

Superstorm Sandy exposure in utero is associated with neurobehavioral phenotypes and brain structure alterations in childhood: A machine learning approach.

Introduction: Prenatal maternal stress (PNMS), including exposure to natural disasters, has been shown to serve as a risk factor for future child psychopathology and suboptimal brain development, particularly among brain regions shown to be sensitive to stress and trauma exposure. However, statistical approaches deployed in most studies are usually constrained by a limited number of variables for the sake of statistical power. Explainable machine learning, on the other hand, enables the study of high data dimension and offers novel insights into the prominent subset of behavioral phenotypes and brain regions most susceptible to PNMS. In the present study, we aimed to identify the most important child neurobehavioral and brain features associated with in utero exposure to Superstorm Sandy (SS). Methods: By leveraging an explainable machine learning technique, the Shapley additive explanations method, we tested the marginal feature effect on SS exposures and examined the individual variable effects on disaster exposure. Results: Results show that certain brain regions are especially sensitive to in utero exposure to SS. Specifically, in utero SS exposure was associated with larger gray matter volume (GMV) in the right caudate, right hippocampus, and left amygdala and smaller GMV in the right parahippocampal gyrus. Additionally, higher aggression scores at age 5 distinctly correlated with SS exposure. Discussion: These findings suggest in utero SS exposure may be associated with greater aggression and suboptimal developmental alterations among various limbic and basal ganglia brain regions.

A new formula to predict the size and insertion depth of cuffed nasotracheal tube in children receiving dental surgery: a retrospective study.

This retrospective study aimed to develop a new formula for selecting the appropriate size and determining the depth of the cuffed nasotracheal intubation (NTI) for a cuffed endotracheal tube (cETT) in pediatric patients undergoing dental surgery. In addition, the clinical data on cETT (i.e., the size and depth of insertion) was compared with those calculated with age-based formulas to evaluate their correlation. A total number of 684 patients who received NTI were enrolled (healthy group, n = 607; special-need group, n = 77). The ETT size used in real-world scenarios was smaller (i.e., about 0.5 and 0.94 mm) than the age-based formula, while the ETT depth was greater (i.e., about 1.5 cm) than the age-based formula in both groups. In the healthy group, age, gender, and body weight were identified as predictors of ETT size and depth through multiple linear regression analysis, while only age and body weight were predictors in the special-needs group. New formulas were developed based on these findings, with ETT size = 3.98 + 0.052 × age + 0.048 × gender (male = 1, female = 0) + 0.023 × body weight (kg) and ETT depth = 15.1 + 0.43 × age + 0.300 × gender (male = 1, female = 0) + 0.007 × body weight (kg). The new formula could be useful for both healthy and special-need pediatric populations undergoing dental procedures.

Changes in the texture and viscoelastic properties of bread containing rice porridge during storage.

The objective of this study was to investigate the effects of rice porridge on the texture and viscoelastic properties of bread during storage. Three types of bread, wheat flour bread, 15% rice flour bread, and 15% rice porridge bread, were prepared. After baking and storing the bread for 24 h, 48 h, and 72 h at room temperature, we measured the texture and viscoelastic properties of the bread crumbs by texture profile analysis (TPA) and creep test. The 15% rice porridge bread showed a significantly higher specific volume and maintained softer crumbs than the other two types (p<0.05). It also had a slightly stickier texture than the others. It can be concluded that rice porridge improves the specific volume, texture, and viscoelastic properties of bread crumbs during storage.

GJB2 mutation spectrum in the Taiwanese population and genotype-phenotype comparisons in patients with hearing loss carrying GJB2 c.109G>A and c.235delC mutations.

Hearing loss, the most common sensory abnormality, is caused by the death of or damage to inner ear hair cells. Genetic mutations are the main cause of hearing loss. We used nex-generation sequencing data released by the Taiwan Biobank to investigate the GJB2 mutation spectrum in 1517 patients. We compared hearing function in Taiwanese patients with nonsyndromic hearing loss (NSHL) caused by the two most common GJB2 mutations c.109G>A (p.V37I) and c.235delC. We extracted DNA from the oral mucosa of patients with NSHL and performed Sanger sequencing to confirm the genotype. Of 240 patients with NSHL, we identified 25 with GJB2 c.109G>A and 9 with GJB2 c.235delC mutations, after excluding patients aged >10 years, in whom hearing loss may have been caused due to age-related degeneration. We investigated genotype-phenotype correlations in patients harboring GJB2 c.109G>A and c.235delC mutations. Furthermore, we described the GJB2 mutation spectrum in the Taiwanese population and identified the role of homozygous and heterozygous GJB2 mutations associated with hearing phenotypes in patients with NSHL. Thus, our study provides insights into the complexity of GJB2 genetics. Our data indicate that GJB2 c.109G>A heterozygotes had poorer hearing than did homozygotes. The mechanism underlying the more severe phenotype in heterozygotes and whether the phenotype is caused by GJB2 heterozygotes or compound heterozygotes warrant future investigation.

The Application of Artificial-Intelligence-Assisted Dental Age Assessment in Children with Growth Delay.

BACKGROUND: This study aimed to reveal the efficacy of the artificial intelligence (AI)-assisted dental age (DA) assessment in identifying the characteristics of growth delay (GD) in children. METHODS: The panoramic films matching the inclusion criteria were collected for the AI model training to establish the population-based DA standard. Subsequently, the DA of the validation dataset of the healthy children and the images of the GD children were assessed by both the conventional methods and the AI-assisted standards. The efficacy of all the studied modalities was compared by the paired sample t-test. RESULTS: The AI-assisted standards can provide much more accurate chronological age (CA) predictions with mean errors of less than 0.05 years, while the traditional methods presented overestimated results in both genders. For the GD children, the convolutional neural network (CNN) revealed the delayed DA in GD children of both genders, while the machine learning models presented so only in the GD boys. CONCLUSION: The AI-assisted DA assessments help overcome the long-standing populational limitation observed in traditional methods. The image feature extraction of the CNN models provided the best efficacy to reveal the nature of delayed DA in GD children of both genders.

Diagnosis of Polypoidal Choroidal Vasculopathy From Fluorescein Angiography Using Deep Learning.

PURPOSE: To differentiate polypoidal choroidal vasculopathy (PCV) from choroidal neovascularization (CNV) and to determine the extent of PCV from fluorescein angiography (FA) using attention-based deep learning networks. METHODS: We build two deep learning networks for diagnosis of PCV using FA, one for detection and one for segmentation. Attention-gated convolutional neural network (AG-CNN) differentiates PCV from other types of wet age-related macular degeneration. Gradient-weighted class activation map (Grad-CAM) is generated to highlight important regions in the image for making the prediction, which offers explainability of the network. Attention-gated recurrent neural network (AG-PCVNet) for spatiotemporal prediction is applied for segmentation of PCV. RESULTS: AG-CNN is validated with a dataset containing 167 FA sequences of PCV and 70 FA sequences of CNV. AG-CNN achieves a classification accuracy of 82.80% at image-level, and 86.21% at patient-level for PCV. Grad-CAM shows that regions contributing to decision-making have on average 21.91% agreement with pathological regions identified by experts. AG-PCVNet is validated with 56 PCV sequences from the EVEREST-I study and achieves a balanced accuracy of 81.132% and dice score of 0.54. CONCLUSIONS: The developed software provides a means of performing detection and segmentation of PCV on FA images for the first time. This study is a promising step in changing the diagnostic procedure of PCV and therefore improving the detection rate of PCV using FA alone. TRANSLATIONAL RELEVANCE: The developed deep learning system enables early diagnosis of PCV using FA to assist the physician in choosing the best treatment for optimal visual prognosis.

Management of Patients with Type V Hyperlipoproteinemia: An Uncommon Phenotype of Dyslipidemia with Chylomicronemia and Severe Hypertriglyceridemia.

Hypertriglyceridemia (HTG) remains a risk-enhancing factor of atherosclerotic cardiovascular disease. We aimed to report real-world data on the management of patients with type V hyperlipoproteinemia (HLP5), an uncommon phenotype of dyslipidemia characterized by fasting chylomicronemia and severe HTG. Between July 2018 and May 2021, 90 patients with HTG, including 83 patients with type IV hyperlipoproteinemia (HLP4) and 7 patients with HLP5, were identified by plasma apolipoprotein B (apoB) and lipoprotein electrophoresis. Patients with HLP5 were younger, had higher total cholesterol (TC) (264.9 ± 26.7 mg/dL vs. 183.9 ± 26.1 mg/dL; p < 0.01) and higher triglyceride (TG) (1296.7 ± 380.5 mg/dL vs. 247.6 ± 96.1 mg/dL; p < 0.01), and had lower high-density lipoprotein cholesterol (HDL-C) (30.6 ± 4.8 mg/dL vs. 40.5 ± 8.7 mg/dL; p < 0.01) and lower low-density lipoprotein cholesterol (LDL-C) (62.9 ± 16.4 vs. 103.0 ± 21.1 mg/dL; p < 0.01) compared with patients with HLP4. Despite an aggressive use of statin and fenofibrate with greater reductions in TG (-65.9 ± 13.7% vs. -27.9 ± 30.5%; p < 0.01) following 6 months of treatment, patients with HLP5 had persistent HTG (440.1 ± 239.0 mg/dL vs. 173.9 ± 94.8 mg/dL; p < 0.01) and an increase in LDL-C (28.3 ± 57.2% vs. -19.5 ± 32.0%; p < 0.01) compared with patients with HLP4. Our findings highlight that the lack of novel TG-lowering medications and management guidelines remains an unmet medical need in patients with HLP5. Closely monitoring lipid profiles, full assessment of individual's risk of cardiovascular disease, and emphasis on medication adherence are of clinical importance.

Targeting protein palmitoylation decreases palmitate­induced sphere formation of human liver cancer cells.

Although non­alcoholic fatty liver disease (NAFLD) is considered a benign disorder, hepatic steatosis has been proposed to be involved in the tumorigenesis of liver cancer. However, the underlying mechanism for carcinogenesis in fatty liver diseases remains unclear. Cancer stem cells (CSCs) have been hypothesized to serve a key role in tumorigenesis. Tumor formation begins with a subset of heterogeneous cells that share properties with stem cells, such as self­renewal and undifferentiated properties. Our previous study reported that the saturated fatty acid palmitate (PA) significantly enhanced the CSC properties of the HepG2 human liver cancer cell line; however, its underlying mechanisms are unknown. In the present study, a proteomic approach was used to investigate the palmitoylation of proteins in HepG2 CSCs. CSC behavior was induced in HepG2 cells via 200 µM PA. Proteomic analysis was performed to identify post­transcriptional modifications of proteins in HepG2 CSCs in response to PA treatment. The present study identified proteins modified by palmitoylation in HepG2 CSC spheres formed following PA treatment. It was therefore hypothesized that palmitoylation may be crucial for CSC sphere formation. Furthermore, the present study demonstrated that two palmitoylation inhibitors, tunicamycin (5, 10 and 25 µg/ml) and 2­bromohexadecanoic acid (25, 50 and 150 µM), significantly decreased CSC sphere formation without affecting cell viability. An association was identified between sphere formation capacity and tumor­initiating capacity of CSCs. The results of the present study demonstrated that protein palmitoylation may influence the PA­induced CSC tumor­initiating capacity, and that the inhibition of palmitoylation may be a suitable chemopreventive strategy for treating patients with NAFLD.