Conformal and conductive biofilm-bridged artificial Z-scheme system for visible light-driven overall water splitting.

Semi-artificial Z-scheme systems offer promising potential toward efficient solar-to-chemical conversion, yet sustainable and stable designs are currently lacking. Here, we developed a sustainable hybrid Z-scheme system capable for visible light-driven overall water splitting by integrating the durability of inorganic photocatalysts with the interfacial adhesion and regenerative property of bacterial biofilms. The Z-scheme configuration is fabricated by drop casting a mixture of photocatalysts onto a glass plate, followed by the growth of biofilms for conformal conductive paste through oxidative polymerization of pyrrole molecules. Notably, the system exhibited scalability indicated by consistent catalytic efficiency across various sheet areas, resistance observed by remarkable maintaining of photocatalytic efficiency across a range of background pressures, and high stability as evidenced by minimal decay of photocatalytic efficiency after 100-hour reaction. Our work thus provides a promising avenue toward sustainable and high-efficiency artificial photosynthesis, contributing to the broader goal of sustainable energy solutions.

Generating Multi-Depth 3D Holograms Using a Fully Convolutional Neural Network.

Efficiently generating 3D holograms is one of the most challenging research topics in the field of holography. This work introduces a method for generating multi-depth phase-only holograms using a fully convolutional neural network (FCN). The method primarily involves a forward-backward-diffraction framework to compute multi-depth diffraction fields, along with a layer-by-layer replacement method (L2RM) to handle occlusion relationships. The diffraction fields computed by the former are fed into the carefully designed FCN, which leverages its powerful non-linear fitting capability to generate multi-depth holograms of 3D scenes. The latter can smooth the boundaries of different layers in scene reconstruction by complementing information of occluded objects, thus enhancing the reconstruction quality of holograms. The proposed method can generate a multi-depth 3D hologram with a PSNR of 31.8 dB in just 90 ms for a resolution of 2160 × 3840 on the NVIDIA Tesla A100 40G tensor core GPU. Additionally, numerical and experimental results indicate that the generated holograms accurately reconstruct clear 3D scenes with correct occlusion relationships and provide excellent depth focusing.

GC/MS-based untargeted metabolomics reveals the differential metabolites for discriminating vintage of Chenxiang-type baijiu.

The "outstanding and unique aged aroma" of Chinese Chenxiang-type baijiu (CXB)-Daoguang 25 (DG25) mainly originates from a "extraordinary storage technology" of Mujiuhai (a wooden container), so it is mysterious and interesting. In this study, an untargeted GC/MS-based metabolomics was used to reveals the volatile differential metabolites for discriminating six different vintages of DG25 combing with chemometrics. A total of 100 volatile metabolites (including unknowns) were extracted and identified, including esters (41%), alcohols (10%) and acids (7%) so on. Finally, 33 differential metabolites were identified as aging-markers. Among them, 25 aging-markers showed a downtrend, including 17 esters such as ethyl acetate, ethyl hexanoate and ethyl palmitate so on. Moreover, it was interesting and to further study that furans showed a significant downtrend. Statistically speaking, ethyl benzoate played an important role in discriminating vintage of 1Y and 3Y, and the other 24 differential metabolites with downtrend discriminating the unstored (0Y-aged) DG25. Eight differential metabolites, such as ethyl octanoate, benzaldehyde, 3-methylbutanol and 1,1-diethoxyaccetal so on increased during aging of DG25, and they played a statistical role in discriminating the 5Y-, 10Y- and 20Y-aged DG25. This study provides a theoretical basis way for the formation mechanism of aging aroma for CXB.

Genetic parameter estimation and molecular foundation of chicken beak shape.

The bird beak is mainly functioned as feeding and attacking, and its shape has extremely important significance for survival and reproduction. In chickens, since beak shape could lead to some disadvantages including pecking and waste of feed, it is important to understand the inheritance of chicken beak shape. In the present study, we firstly established 4 indicators to describe the chicken beak shapes, including upper beak length (UL), lower beak length (LL), distance between upper and lower beak tips (DB) and upper beak curvature (BC). And then, we measured the 4 beak shape indicators as well as some production traits including body weight (BW), shank length (SL), egg weight (EW), eggshell strength (ES) of a layer breed, Rhode Island Red (RIR), in order to estimate genetic parameters of chicken beak shape. The heritabilities of UL and LL were 0.41 and 0.37, and the heritabilities of DB and BC were 0.22 and 0.21, indicating that beak shape was a highly or mediumly heritable. There were significant positive genetic and phenotypic correlations among UL, LL, and DB. And UL was positively correlated with body weight (BW18) and shank length (SL18) at 18 weeks of age in genetics, and DB was positively correlated with BC in terms of genetics and phenotype. We also found that layers of chicken cages played a role on beak shape, which could be attributed to the difference of lightness in different cage layers. By a genome-wide association study (GWAS) for the chicken UL, we identified 9 significant candidate genes associated with UL in RIR. For the variants with low minor allele frequencies (MAF <0.01) and outside of high linkage disequilibrium (LD) regions, we also conducted rare variant association studies (RVA) and GWAS to find the association between genotype and phenotype. We also analyzed transcriptomic data from multiple tissues of chicken embryos and revealed that all of the 9 genes were highly expressed in beak of chicken embryos, indicating their potential function for beak development. Our results provided the genetic foundation of chicken beak shape, which could help chicken breeding on beak related traits.

Comparison of the efficacy and tolerability of different repetitive transcranial magnetic stimulation modalities for post-stroke dysphagia: a systematic review and Bayesian network meta-analysis protocol.

INTRODUCTION: Up to 78% of patients who had a stroke develop post-stroke dysphagia (PSD), a significant consequence. Life-threatening aspiration pneumonia, starvation, and water and electrolyte abnormalities can result. Several meta-analyses have shown that repeated transcranial magnetic stimulation (rTMS) improves swallowing in patients who had a stroke; however, the optimum model is unknown. This study will be the first Bayesian network meta-analysis (NMA) to determine the best rTMS modalities for swallowing of patients who had a stroke. METHODS AND ANALYSIS: PubMed, Web of Science, Embase, Google Scholar, Cochrane, the Chinese National Knowledge Infrastructure, the Chongqing VIP Database and WanFang Data will be searched from their creation to 2 September 2023. All randomised controlled trials associated with rTMS for PSD will be included. Only Chinese or English results will be studied. Two researchers will independently review the literature and extract data, then use the Cochrane Collaboration's Risk of Bias 2.0 tool to assess the included studies' methodological quality. The primary outcome is swallowing function improvement, whereas secondary outcomes include side effects (eg, paraesthesia, vertigo, seizures) and quality of life. A pairwise meta-analysis and NMA based on a Bayesian framework will be conducted using Stata and R statistical software. The Grading of Recommendations Assessment, Development, and Evaluation system will assess outcome indicator evidence quality. ETHICS AND DISSEMINATION: As all data in this study will be taken from the literature, ethical approval is not needed. We will publish our work in peer-reviewed publications and present it at academic conferences. PROSPERO REGISTRATION NUMBER: CRD42023456386.

Multimodal Omics Approaches to Aging and Age-Related Diseases.

Aging is associated with a progressive decline in physiological capacities and an increased risk of aging-associated disorders. An increasing body of experimental evidence shows that aging is a complex biological process coordinately regulated by multiple factors at different molecular layers. Thus, it is difficult to delineate the overall systematic aging changes based on single-layer data. Instead, multimodal omics approaches, in which data are acquired and analyzed using complementary omics technologies, such as genomics, transcriptomics, and epigenomics, are needed for gaining insights into the precise molecular regulatory mechanisms that trigger aging. In recent years, multimodal omics sequencing technologies that can reveal complex regulatory networks and specific phenotypic changes have been developed and widely applied to decode aging and age-related diseases. This review summarizes the classification and progress of multimodal omics approaches, as well as the rapidly growing number of articles reporting on their application in the field of aging research, and outlines new developments in the clinical treatment of age-related diseases based on omics technologies.

Genetic parameter estimation and molecular foundation of chicken egg-laying trait.

The age of first egg (AFE) in chicken can affect early and even life-time egg production performance to some extent, and therefore is an important economic trait that affects production efficiency. To better understand the genetic patterns of AFE and other production traits including body weight at first egg (BWA), first egg weight (FEW), and total egg number from AFE to 58 wk of age (total-EN), we recorded the production performance of 2 widely used layer breeds, white leghorn (WL) and Rhode Island Red (RIR) and estimated genetic parameters based on pedigree and production data. The results showed that the heritability of AFE in both breeds ranged from 0.4 to 0.6, and AFE showed strong positive genetic and phenotypic correlations to BWA as well as FEW, while showing strong negative genetic and phenotypic correlations with total-EN. Furtherly, by genome-wide association analysis study (GWAS), we identified 12 and 26 significant SNPs to be related to AFE in the 2-layer breeds, respectively. A total of 18 genes were identified that could affect AFE based on the significant SNP annotations obtained, but there were no gene overlapped in the 2 breeds indicating the genetic foundation of AFE could differ from breed to breed. Our results provided a deeper understanding of genetic patterns and molecular basement of AFE in different breeds and could help in the selection of egg production traits.

Molecular genetic foundation of a sex-linked tailless trait in Hongshan chicken by whole genome data analysis.

As a Chinese local chicken breed, Hongshan chickens have 2 kinds of tail feather phenotypes, normal and taillessness. Our previous studies showed that taillessness was a sex-linked dominant trait. Abnormal development of the tail vertebrae could be explained this phenomenon in some chicken breeds. However, the number of caudal vertebrae in rumpless Hongshan chickens was normal, so rumplessness in Hongshan chicken was not related to the development of the caudal vertebrae. Afterwards, we found that rumplessness in Hongshan was due to abnormal development of tail feather rather than abnormal development of caudal vertebrae. In order to understand the genetic foundation of the rumplessness of Hongshan chickens, we compared and reanalyzed 2 sets of data in normal and rumpless Hongshan chickens from our previous studies. By joint analysis of genome-wide selection signature analysis and genome-wide association approach, we found that 1 overlapping gene (EDIL3) and 16 peak genes (ENSGALG00000051843, ENSGALG00000053498, ENSGALG00000054800, KIF27, PTPRD, ENSGALG00000047579, ENSGALG00000041052, ARHGEF28, CAMK4, SERINC5, ENSGALG00000050776, ERCC8, MCC, ADAMTS19, ENSGALG00000053322, CHRNA8) located on the Z chromosome was associated with the rumpless trait. The results of this study furtherly revealed the molecular mechanism of the rumpless trait in Hongshan chickens, and identified the candidate genes associated with this trait. Our results will help to improve the shape of chicken tail feathers and to rise individual economic value in some specific market in China.

The FBXW7-binding sites on FAM83D are potential targets for cancer therapy.

Increasing evidence shows the oncogenic function of FAM83D in human cancer, but how FAM83D exerts its oncogenic function remains largely unclear. Here, we investigated the importance of FAM83D/FBXW7 interaction in breast cancer (BC). We systematically mapped the FBXW7-binding sites on FAM83D through a comprehensive mutational analysis together with co-immunoprecipitation assay. Mutations at the FBXW7-binding sites on FAM83D led to that FAM83D lost its capability to promote the ubiquitination and proteasomal degradation of FBXW7; cell proliferation, migration, and invasion in vitro; and tumor growth and metastasis in vivo, indicating that the FBXW7-binding sites on FAM83D are essential for its oncogenic functions. A meta-evaluation of FAM83D revealed that the prognostic impact of FAM83D was independent on molecular subtypes. The higher expression of FAM83D has poorer prognosis. Moreover, high expression of FAM83D confers resistance to chemotherapy in BCs, which is experimentally validated in vitro. We conclude that identification of FBXW7-binding sites on FAM83D not only reveals the importance for FAM83D oncogenic function, but also provides valuable insights for drug target.

Fully Bio-Based 2,5-Furandicarboxylic Acid Polyester toward Plastics with Mechanically Robust, Excellent Gas Barrier and Fast Degradation.

Aliphatic-aromatic copolyesters offer a promising solution to mitigate plastic pollution, but high content of aliphatic units (>40 %) often suffer from diminished comprehensive performances. Poly(butylene oxalate-co-furandicarboxylate) (PBOF) copolyesters were synthesized by precisely controlling the oxalic acid content from 10 % to 60 %. Compared with commercial PBAT, the barrier properties of PBOF for H2O and O2 increased by more than 6 and 26 times, respectively. The introduction of the oxalic acid units allowed the water contact angle to be reduced from 82.5° to 62.9°. Superior hydrophilicity gave PBOF an excellent degradation performance within a 35-day hydrolysis. Interestingly, PBO20F and PBO30F also displayed obvious decrease of molecular weight during hydrolysis, with elastic modulus >1 GPa and tensile strength between 35-54 MPa. PBOF achieved the highest hydrolysis rates among the reported PBF-based copolyesters. The hydrolytic mechanism was further explored based on Fukui function analysis and density functional theory (DFT) calculation. Noncovalent analysis indicated that the water molecules formed hydrogen bonding interaction with adjacent ester groups and thus improved the reactivity of carbonyl carbon. PBOF not only meet the requirements of the high-performance packaging market but can quickly degrade after the end of their usage cycles, providing a new choice for green and environmental protection.

Combination Nano-Delivery Systems Remodel the Immunosuppressive Tumor Microenvironment for Metastatic Triple-Negative Breast Cancer Therapy.

Triple-negative breast cancer (TNBC) is an aggressive type of breast cancer for which effective therapies are lacking. Targeted remodeling of the immunosuppressive tumor microenvironment (TME) and activation of the body's immune system to fight tumors with well-designed nanoparticles have emerged as pivotal breakthroughs in tumor treatment. To simultaneously remodel the immunosuppressive TME and trigger immune responses, we designed two potential therapeutic nanodelivery systems to inhibit TNBC. First, the bromodomain-containing protein 4 (BRD4) inhibitor JQ1 and the cyclooxygenase-2 (COX-2) inhibitor celecoxib (CXB) were coloaded into chondroitin sulfate (CS) to obtain CS@JQ1/CXB nanoparticles (NPs). Then, the biomimetic nanosystem MM@P3 was prepared by coating branched polymer poly(ß-amino ester) self-assembled NPs with melittin embedded macrophage membranes (MM). Both in vitro and in vivo, the CS@JQ1/CXB and MM@P3 NPs showed excellent immune activation efficiencies. Combination treatment exhibited synergistic cytotoxicity, antimigration ability, and apoptosis-inducing and immune activation effects on TNBC cells and effectively suppressed tumor growth and metastasis in TNBC tumor-bearing mice by activating the tumor immune response and inhibiting angiogenesis. In summary, this study offers a novel combinatorial immunotherapeutic strategy for the clinical TNBC treatment.

The sirtuin-associated human senescence program converges on the activation of placenta-specific gene PAPPA.

Sirtuins are pro-longevity genes with chromatin modulation potential, but how these properties are connected is not well understood. Here, we generated a panel of isogeneic human stem cell lines with SIRT1-SIRT7 knockouts and found that any sirtuin deficiency leads to accelerated cellular senescence. Through large-scale epigenomic analyses, we show how sirtuin deficiency alters genome organization and that genomic regions sensitive to sirtuin deficiency are preferentially enriched in active enhancers, thereby promoting interactions within topologically associated domains and the formation of de novo enhancer-promoter loops. In all sirtuin-deficient human stem cell lines, we found that chromatin contacts are rewired to promote aberrant activation of the placenta-specific gene PAPPA, which controls the pro-senescence effects associated with sirtuin deficiency and serves as a potential aging biomarker. Based on our survey of the 3D chromatin architecture, we established connections between sirtuins and potential target genes, thereby informing the development of strategies for aging interventions.

New modifications of PBAT by a small amount of oxalic acid: Fast crystallization and enhanced degradation in all natural environments.

Biodegradable plastics are often mistakenly thought to be capable of degrading in any environment, but their slow degradation rate in the natural environment is still unsatisfactory. We synthetized a novel series of poly(butylene oxalate-co-adipate-co-terephthalate) (PBOAT) with unchanged melting point (135 °C), high elastic modulus (140 - 219 MPa) and elongation at break (478 - 769%). Fast isothermal crystallization with a semi-crystallization time < 20 s was demonstrated by the PBOAT. In N2 and air atmospheres, the PBOAT maintained the Td,5% higher than 329 °C. They also had good thermal stability at melt processing temperature for more than 20 min. PBOAT exhibited faster hydrolysis and seawater degradation, even under natural soil burial without light, but still kept stable under low humidity conditions during the storage and the shelf-life. Moreover, the hydrolysis mechanisms were clarified based on Fukui function analysis and DFT calculation, indicating that the hydrolysis of PBOAT would be more straightforward. The mechanism of soil burial is also elucidated through detailed characterization of the structure changes. The PBOAT offered a fresh approach to the development of high-performing, naturally degradable materials.

Association between polyunsaturated fatty acids and progression among patients with diabetic kidney disease.

AIMS: Diabetic kidney disease (DKD) is the major complication of diabetes mellitus (DM) and one of the leading causes of end-stage renal disease. Early detection and treatment are contributing to delay the progression of DKD. Dietary management has potential benefits for DKD, especially the intake of polyunsaturated fatty acids (PUFAs). However, there is a lack of sufficient evidence, so we aimed to explore the association between PUFAs intake and DKD progression. METHODS: In the National Heath and Nutrition Examination Survey (NHANES) between 2011-2018, a cross-sectional study was conducted among adults with T2DM. DKD was diagnosed with urine albumin to creatinine ratio (ACR) ≥ 30 mg/g or estimated glomerular filtration rate (eGFR) ＜60 ml/min/1.73 m2. Using Survey package of R to arrange the collected PUFAs intake data in order from small to large and divide them into four equal parts, which were expressed as Q1, Q2, Q3 and Q4 respectively. To investigate the association between PUFAs intake and DKD, a weighted univariate logistic regression analysis was performed and the odds ratio (OR) and 95% confidence interval (CI) were calculated for the association with DKD and PUFAs quartiles. RESULTS: The study involved 3287 participants with T2DM, including 2043 non-DKD and 1244 DKD patients. The results showed that the intake of PUFAs was a protective factor for DKD (p = 0.022), and with the increase of the PUFAs, renal function improved in DKD patients, the adjusted mean of eGFR and Scr changing from 57 (41, 86) in Q1 to 71 (55, 101) ml/min in Q4 (p 0.001), 103 (73, 131) in Q1 to 90 (68, 117) in Q4 (p = 0.031), respectively. CONCLUSION: Our study indicated that intake of more PUFAs may contribute to delay DKD progression, while different n-6/n-3 ratios need to be explored to protect the kidney.

Detection of Treatment Response in Triple-Negative Breast Tumors to Paclitaxel Using MRI Cell Size Imaging.

BACKGROUND: Breast cancer treatment response evaluation using the response evaluation criteria in solid tumors (RECIST) guidelines, based on tumor volume changes, has limitations, prompting interest in novel imaging markers for accurate therapeutic effect determination. PURPOSE: To use MRI-measured cell size as a new imaging biomarker for assessing chemotherapy response in breast cancer. STUDY TYPE: Longitudinal; animal model. STUDY POPULATION: Triple-negative human breast cancer cell (MDA-MB-231) pellets (4 groups, n = 7) treated with dimethyl sulfoxide (DMSO) or 10 nM of paclitaxel for 24, 48, and 96 hours, and 29 mice with MDA-MB-231 tumors in right hind limbs treated with paclitaxel (n = 16) or DMSO (n = 13) twice weekly for 3 weeks. FIELD STRENGTH/SEQUENCE: Oscillating gradient spin echo and pulsed gradient spin echo sequences at 4.7 T. ASSESSMENT: MDA-MB-231 cells were analyzed using flowcytometry and light microscopy to assess cell cycle phases and cell size distribution. MDA-MB-231 cell pellets were MR imaged. Mice were imaged weekly, with 9, 6, and 14 being sacrificed for histology after MRI at weeks 1, 2, and 3, respectively. Microstructural parameters of tumors/cell pellets were derived by fitting diffusion MRI data to a biophysical model. STATISTICAL TESTS: One-way ANOVA compared cell sizes and MR-derived parameters between treated and control samples. Repeated measures 2-way ANOVA with Bonferroni post-tests compared temporal changes in MR-derived parameters. A P-value <0.05 was considered statistically significant. RESULTS: In vitro experiments showed that the mean MR-derived cell sizes of paclitaxel-treated cells increased significantly with a 24-hours treatment and decreased (P = 0.06) with a 96-hour treatment. For in vivo xenograft experiments, the paclitaxel-treated tumors showed significant decreases in cell size at later weeks. MRI observations were supported by flowcytometry, light microscopy, and histology. DATA CONCLUSIONS: MR-derived cell size may characterize the cell shrinkage during treatment-induced apoptosis, and may potentially provide new insights into the assessment of therapeutic response. LEVEL OF EVIDENCE: 2 TECHNICAL EFFICACY STAGE: 4.

Human ESC-derived vascular cells promote vascular regeneration in a HIF-1α dependent manner.

Hypoxia-inducible factor (HIF-1α), a core transcription factor responding to changes in cellular oxygen levels, is closely associated with a wide range of physiological and pathological conditions. However, its differential impacts on vascular cell types and molecular programs modulating human vascular homeostasis and regeneration remain largely elusive. Here, we applied CRISPR/Cas9-mediated gene editing of human embryonic stem cells and directed differentiation to generate HIF-1α-deficient human vascular cells including vascular endothelial cells, vascular smooth muscle cells, and mesenchymal stem cells (MSCs), as a platform for discovering cell type-specific hypoxia-induced response mechanisms. Through comparative molecular profiling across cell types under normoxic and hypoxic conditions, we provide insight into the indispensable role of HIF-1α in the promotion of ischemic vascular regeneration. We found human MSCs to be the vascular cell type most susceptible to HIF-1α deficiency, and that transcriptional inactivation of ANKZF1, an effector of HIF-1α, impaired pro-angiogenic processes. Altogether, our findings deepen the understanding of HIF-1α in human angiogenesis and support further explorations of novel therapeutic strategies of vascular regeneration against ischemic damage.

Effect of robot-assisted gait training on motor dysfunction in Parkinson's patients:A systematic review and meta-analysis.

BACKGROUND: Robot-assisted gait training (RAGT) has been reported to treat motor dysfunction in patients with Parkinson's disease (PD) in the last few years. However, the benefits of RAGT for treating motor dysfunction in PD are still unclear. OBJECTIVES: To investigate the efficacy of RAGT for motor dysfunction in PD patients. METHODS: We searched PubMed, Web of Science, Cochrane Library, Embase, CNKI, Wanfang, Chinese Biomedical Literature Database (CBM), and Chinese VIP Database for randomized controlled trials investigating RAGT to improve motor dysfunction in PD from the databases' inception dates until September 1, 2022. The following outcome indexes were employed to evaluate motor dysfunction: the Berg Balance Scale (BBS), Activities-specific Balance Confidence Scale (ABC), 10-Meter Walk Test gait speed (10-MWT), gait speed, stride length, cadence Unified Parkinson Disease Rating Scale Part III (UPDRS III), 6-Minute Walk Test (6MWT), and the Timed Up and Go test (TUG). The meta-analysis was performed using the proper randomeffect model or fixed-effect model to evaluate the difference in efficacy between the RAGT and the control groups. The Cochrane Risk of Bias Tool was used for the included studies and Grading of Recommendations, Assessment, Development, and Evaluations (GRADE) was used to interpret the certainty of the results. RESULTS: The results consisted of 17 studies comprising a total of 670 participants. Six hundred and seven PD patients with motor dysfunction were included: 335 in the RAGT group and 335 in the control group. This meta-analysis results established that when compared with the control group, robot-assisted gait training improved the BBS results of PD patients (MD: 2.80, 95%CI: 2.11-3.49, P< 0.00001), ABC score (MD: 7.30, 95%CI: 5.08-9.52, P< 0.00001), 10-MWT (MD: 0.06, 95%CI: 0.03-0.10, P= 0.0009), gait speed (MD: 3.67, 95%CI: 2.58-4.76, P< 0.00001), stride length (MD: 5.53, 95%CI: 3.64-7.42, P< 0.00001), cadence (MD: 4.52, 95%CI: 0.94-8.10, P= 0.01), UPDRS III (MD: -2.16, 95%CI: -2.48--1.83, P< 0.00001), 6MWT (MD: 13.87, 95%CI: 11.92-15.82, P< 0.00001). However, RAGT did not significantly improve the TUG test result of patients with PD (MD =-0.56, 95% CI: -1.12-0.00, P= 0.05). No safety concerns or adverse reactions among robot-assisted gait training patients were observed. CONCLUSION: Even though RAGT can improve balance function, walking function, and gait performance and has demonstrated positive results in several studies, there is currently insufficient compelling evidence to suggest that it can improve all aspects of lower motor function.

Clinical evaluation of deep learning systems for assisting in the diagnosis of the epiretinal membrane grade in general ophthalmologists.

BACKGROUND: Epiretinal membrane (ERM) is a common age-related retinal disease detected by optical coherence tomography (OCT), with a prevalence of 34.1% among people over 60 years old. This study aims to develop artificial intelligence (AI) systems to assist in the diagnosis of ERM grade using OCT images and to clinically evaluate the potential benefits and risks of our AI systems with a comparative experiment. METHODS: A segmentation deep learning (DL) model that segments retinal features associated with ERM severity and a classification DL model that grades the severity of ERM were developed based on an OCT dataset obtained from three hospitals. A comparative experiment was conducted to compare the performance of four general ophthalmologists with and without assistance from the AI in diagnosing ERM severity. RESULTS: The segmentation network had a pixel accuracy (PA) of 0.980 and a mean intersection over union (MIoU) of 0.873, while the six-classification network had a total accuracy of 81.3%. The diagnostic accuracy scores of the four ophthalmologists increased with AI assistance from 81.7%, 80.7%, 78.0%, and 80.7% to 87.7%, 86.7%, 89.0%, and 91.3%, respectively, while the corresponding time expenditures were reduced. The specific results of the study as well as the misinterpretations of the AI systems were analysed. CONCLUSION: Through our comparative experiment, the AI systems proved to be valuable references for medical diagnosis and demonstrated the potential to accelerate clinical workflows. Systematic efforts are needed to ensure the safe and rapid integration of AI systems into ophthalmic practice.