DeepQuality improves infant retinopathy screening.

Image quality variation is a prominent cause of performance degradation for intelligent disease diagnostic models in clinical applications. Image quality issues are particularly prominent in infantile fundus photography due to poor patient cooperation, which poses a high risk of misdiagnosis. Here, we developed a deep learning-based image quality assessment and enhancement system (DeepQuality) for infantile fundus images to improve infant retinopathy screening. DeepQuality can accurately detect various quality defects concerning integrity, illumination, and clarity with area under the curve (AUC) values ranging from 0.933 to 0.995. It can also comprehensively score the overall quality of each fundus photograph. By analyzing 2,015,758 infantile fundus photographs from real-world settings using DeepQuality, we found that 58.3% of them had varying degrees of quality defects, and large variations were observed among different regions and categories of hospitals. Additionally, DeepQuality provides quality enhancement based on the results of quality assessment. After quality enhancement, the performance of retinopathy of prematurity (ROP) diagnosis of clinicians was significantly improved. Moreover, the integration of DeepQuality and AI diagnostic models can effectively improve the model performance for detecting ROP. This study may be an important reference for the future development of other image-based intelligent disease screening systems.

Accuracy of the FY-L formula in calculating intraocular lens power after small-incision lenticule extraction.

Purpose: The study aimed to assess the accuracy of the FY-L formula in calculating intraocular lens (IOL) power after small-incision lenticule extraction (SMILE). Methods: For the post-SMILE IOL calculation of the same eye, the IOL power targeting the pre-SMILE eyes' lowest myopic refractive error was used. The FY-L formula, the Emmetropia Verifying Optical Formula (EVO-L), the Barrett True-K no history, and the Shammas-L, respectively, were used to calculate the predicted refractive error of target IOL power. A comparison was made between the change in spherical equivalent induced by SMILE (SMILE-Dif) and the variance between IOL-Dif (IOL-Induced Refractive Error) before and after SMILE. The prediction error (PE) was defined as SMILE-Dif minus IOL-Dif. The proportion of eyes with PEs within ±0.25 D, ±0.50 D, ±0.75 D, and ±1.00 D, the numerical and absolute prediction errors (PEs and AEs), and the median absolute error (MedAE) were compared. Results: In total, 80 eyes from 42 patients who underwent SMILE were included in the study. The FY-L formula generated the sample's lowest mean PE (0.06 ± 0.76 D), MAE (0.58 ± 0.50 D), and MedAE (0.47 D), respectively. The PEs in ±0.25 D, ±0.50 D, ±0.75 D, and ±1.00 D comprised 28.8%, 46.3%, 70.0%, and 87.5%, respectively, for the FY-L formula. Compared to other formulas, the FY-L formula produced the highest value with PEs for the percentage of eyes in ±0.50 D, ±0.75 D, and ±1.00 D. Conclusion: This study demonstrates that the FY-L formula provides satisfactory outcomes in estimating the IOL power in the eyes after SMILE.

ERF5 enhances protocorm-like body regeneration via enhancement of STM expression in Dendrobium orchid.

Orchid plants develop protocorms upon germination and produce protocorm-like structures called protocorm-like bodies (PLBs) from protocorms and somatic cells via tissue culture. Protocorm-like bodies have broad technical application potential in the orchid industry and their regeneration is a distinct developmental process in the plant kingdom. However, little is known about this unparalleled developmental program. In this study, we identified a PLB-abundant gene, ethylene response factor (ERF), and a transcription factor named DoERF5, and determined its important role in PLB regeneration in Dendrobium orchid. Overexpression of DoERF5 in Dendrobium greatly enhanced the PLB regeneration from PLB and stem explants, and upregulated the expression of WOUND-INDUCED DEDIFFERENTIATION (DoWIND) homologs and SHOOT MERISTEMLESS (DoSTM), as well as the genes involved in cytokinin biosynthesis (DoIPT) and the cytokinin response factors (DoARRs). However, silencing DoERF5 reduced the regeneration rate of PLBs, and downregulated the expression of DoWIND homologs, DoSTM and DoARRs. We demonstrated that DoERF5 is directly bound to the DoSTM promoter and regulates its expression. In addition, overexpression of DoSTM in Dendrobium orchid resulted in favorable regeneration of PLBs. Our results clarify that DoERF5 regulates the regeneration of PLB by enhancing DoSTM expression. Our findings provide new insights into how DoERF5 mediates PLB regeneration and offers technical potential in improving clonal propagation, preservation, and the bioengineering of orchids.

Genome-wide identification of Aux/IAA and ARF gene families reveal their potential roles in flower opening of Dendrobium officinale.

BACKGROUND: The auxin indole-3-acetic acid (IAA) is a vital phytohormone that influences plant growth and development. Our previous work showed that IAA content decreased during flower development in the medicinally important orchid Dendrobium officinale, while Aux/IAA genes were downregulated. However, little information about auxin-responsive genes and their roles in D. officinale flower development exists. RESULTS: This study validated 14 DoIAA and 26 DoARF early auxin-responsive genes in the D. officinale genome. A phylogenetic analysis classified the DoIAA genes into two subgroups. An analysis of cis-regulatory elements indicated that they were related by phytohormones and abiotic stresses. Gene expression profiles were tissue-specific. Most DoIAA genes (except for DoIAA7) were sensitive to IAA (10 µmol/L) and were downregulated during flower development. Four DoIAA proteins (DoIAA1, DoIAA6, DoIAA10 and DoIAA13) were mainly localized in the nucleus. A yeast two-hybrid assay showed that these four DoIAA proteins interacted with three DoARF proteins (DoARF2, DoARF17, DoARF23). CONCLUSIONS: The structure and molecular functions of early auxin-responsive genes in D. officinale were investigated. The DoIAA-DoARF interaction may play an important role in flower development via the auxin signaling pathway.

Uncovering the involvement of DoDELLA1-interacting proteins in development by characterizing the DoDELLA gene family in Dendrobium officinale.

BACKGROUND: Gibberellins (GAs) are widely involved in plant growth and development. DELLA proteins are key regulators of plant development and a negative regulatory factor of GA. Dendrobium officinale is a valuable traditional Chinese medicine, but little is known about D. officinale DELLA proteins. Assessing the function of D. officinale DELLA proteins would provide an understanding of their roles in this orchid's development. RESULTS: In this study, the D. officinale DELLA gene family was identified. The function of DoDELLA1 was analyzed in detail. qRT-PCR analysis showed that the expression levels of all DoDELLA genes were significantly up-regulated in multiple shoots and GA3-treated leaves. DoDELLA1 and DoDELLA3 were significantly up-regulated in response to salt stress but were significantly down-regulated under drought stress. DoDELLA1 was localized in the nucleus. A strong interaction was observed between DoDELLA1 and DoMYB39 or DoMYB308, but a weak interaction with DoWAT1. CONCLUSIONS: In D. officinale, a developmental regulatory network involves a close link between DELLA and other key proteins in this orchid's life cycle. DELLA plays a crucial role in D. officinale development.

Early detection of visual impairment in young children using a smartphone-based deep learning system.

Early detection of visual impairment is crucial but is frequently missed in young children, who are capable of only limited cooperation with standard vision tests. Although certain features of visually impaired children, such as facial appearance and ocular movements, can assist ophthalmic practice, applying these features to real-world screening remains challenging. Here, we present a mobile health (mHealth) system, the smartphone-based Apollo Infant Sight (AIS), which identifies visually impaired children with any of 16 ophthalmic disorders by recording and analyzing their gazing behaviors and facial features under visual stimuli. Videos from 3,652 children (≤48 months in age; 54.5% boys) were prospectively collected to develop and validate this system. For detecting visual impairment, AIS achieved an area under the receiver operating curve (AUC) of 0.940 in an internal validation set and an AUC of 0.843 in an external validation set collected in multiple ophthalmology clinics across China. In a further test of AIS for at-home implementation by untrained parents or caregivers using their smartphones, the system was able to adapt to different testing conditions and achieved an AUC of 0.859. This mHealth system has the potential to be used by healthcare professionals, parents and caregivers for identifying young children with visual impairment across a wide range of ophthalmic disorders.

Characterization of YABBY genes in Dendrobium officinale reveals their potential roles in flower development.

These YABBY genes are transcription factors (TFs) that play crucial roles in various developmental processes in plants. There is no comprehensive characterization of YABBY genes in a valuable Chinese orchid herb, Dendrobium officinale. In this study, a total of nine YABBY genes were identified in the D. officinale genome. These YABBY genes were divided into four subfamilies: CRC/DL, FIL, INO, and YAB2. Expression pattern analyses showed that eight of the YABBY genes were strongly expressed in reproductive organs (flower buds) but weakly expressed in vegetative organs (roots, leaves, and stems). DoYAB1, DoYAB5, DoDL1, and DoDL3 were abundant in the small flower bud stage, while DoDL2 showed no changes throughout flower development. In addition, DoDL1-3 genes were strongly expressed in the column, tenfold more than in sepals, petals, and the lip. DoYAB1 from the FIL subfamily, DoYAB2 from the YAB2 subfamily, DoYAB3 from the INO subfamily, and DoDL2 and DoDL3 from the CRC/DL subfamily were selected for further analyses. Subcellular localization analysis showed that DoYAB1-3, DoDL2, and DoDL3 were localized in the nucleus. DoYAB2 and DoYAB3 interacted strongly with DoWOX2 and DoWOX4, while DoYAB1 showed a weak interaction with DoWOX4. These results reveal a regulatory network involving YABBY and WOX proteins in D. officinale. Our data provide clues to understanding the role of YABBY genes in the regulation of flower development in this orchid and shed additional light on the function of YABBY genes in plants.

Comparison of robot-assisted vitreoretinal surgery and manual surgery in different preclinical settings: a randomized trial.

Background: Progress in the field of surgical robotics has the potential to allow surgeons to reduce the limitations of human hands and has substantially improved the dexterity and accuracy of surgery. This study aimed to compare robot-assisted vitreoretinal surgery with manual surgery in a simulated setting involving donor porcine eyes. Methods: A randomized controlled trial was conducted involving 10 experienced vitreoretinal experts and 10 residents. Participants were randomized to start with either manual or robot-assisted surgery (n=5 in each group). Participants completed 3 tests consisting of 2 vitreoretinal modules on a virtual-reality simulator and microcannulation in donor porcine eyes. The primary outcome measures were as follows: test completion time, surgical accuracy and tremor control provided by the simulator, and the feasibility of microcannulation in donor porcine eyes. Results: Robot-assisted surgery supported better accuracy and tremor control than manual surgery for vitreoretinal experts (P=0.028 and P=0.002, respectively) and residents (P=0.025 and P<0.001, respectively). Residents improved their microcannulation performance with the assistance of the robot (P=0.038) to a level comparable to that of experts (P=0.49). Robot-assisted surgery was less time-efficient than manual surgery for both residents (P<0.001) and vitreoretinal experts (P<0.001). Conclusions: Compared with manual surgery, robot-assisted vitreoretinal surgery improves the stability of instrument handling and the accuracy of surgery. Robot-assisted surgeries have the potential to shorten the learning curve for residents and improve surgical performance.

A digital mask to safeguard patient privacy.

The storage of facial images in medical records poses privacy risks due to the sensitive nature of the personal biometric information that can be extracted from such images. To minimize these risks, we developed a new technology, called the digital mask (DM), which is based on three-dimensional reconstruction and deep-learning algorithms to irreversibly erase identifiable features, while retaining disease-relevant features needed for diagnosis. In a prospective clinical study to evaluate the technology for diagnosis of ocular conditions, we found very high diagnostic consistency between the use of original and reconstructed facial videos (κ ≥ 0.845 for strabismus, ptosis and nystagmus, and κ = 0.801 for thyroid-associated orbitopathy) and comparable diagnostic accuracy (P ≥ 0.131 for all ocular conditions tested) was observed. Identity removal validation using multiple-choice questions showed that compared to image cropping, the DM could much more effectively remove identity attributes from facial images. We further confirmed the ability of the DM to evade recognition systems using artificial intelligence-powered re-identification algorithms. Moreover, use of the DM increased the willingness of patients with ocular conditions to provide their facial images as health information during medical treatment. These results indicate the potential of the DM algorithm to protect the privacy of patients' facial images in an era of rapid adoption of digital health technologies.

Transcriptomic and metabolomic analyses reveal the main metabolites in Dendrobium officinale leaves during the harvesting period.

Dendrobium officinale, which is a medicine food homology plant, contains many metabolites, especially polysaccharides and flavonoids. Unlike flowers and stems, which are the most frequently harvested organs for a variety of uses, leaves tend to be discarded. This study assessed main metabolites in leaves to identify the most appropriate timing of collection during harvest, which was divided into three stages (S1-S3: 8, 10, and 11 months after sprouting, respectively). Metabolomic and transcriptomic analyses of S1-S3 were performed. Water-soluble polysaccharides (WSPs), flavonoids and free amino acids (FAAs) were detected in leaves. WSPs decreased from S1 to S3 but flavonoids and some FAAs (e.g., phophoserine) increased from S1 to S2, then decreased from S2 to S3. In all three stages, mannose was the dominant monosaccharide among WSPs, followed by glucose. In S2, 35 flavonoids were identified, the most abundant being rutin, schaftoside and vitexin, while 34 FAAs were identified in all three stages, the most abundant being tyrosine, phosphoserine and alanine. A total of 2584, 3414 and 2032 differentially expressed genes (DEGs) were discovered in S1 vs S2, S1 vs S3 and S1 vs S3, respectively. Correlation analysis revealed that five DEGs (DoSUS, DoXYLA, DoFRK, DoGMP, and DoCSLA), two DEGs (DoDFR, and DoANS) and a single DEG (DoPGAM) were involved in the metabolism of WSPs, flavonoids and phosphoserine, respectively. The findings of this study lay a foundation for the commercial exploitation of metabolites in the harvested leaves of D. officinale, and the use of detected DEGs in applied genetic studies.

Genome-Wide Identification and Analysis of the APETALA2 (AP2) Transcription Factor in Dendrobium officinale.

The APETALA2 (AP2) transcription factors (TFs) play crucial roles in regulating development in plants. However, a comprehensive analysis of the AP2 family members in a valuable Chinese herbal orchid, Dendrobium officinale, or in other orchids, is limited. In this study, the 14 DoAP2 TFs that were identified from the D. officinale genome and named DoAP2-1 to DoAP2-14 were divided into three clades: euAP2, euANT, and basalANT. The promoters of all DoAP2 genes contained cis-regulatory elements related to plant development and also responsive to plant hormones and stress. qRT-PCR analysis showed the abundant expression of DoAP2-2, DoAP2-5, DoAP2-7, DoAP2-8 and DoAP2-12 genes in protocorm-like bodies (PLBs), while DoAP2-3, DoAP2-4, DoAP2-6, DoAP2-9, DoAP2-10 and DoAP2-11 expression was strong in plantlets. In addition, the expression of some DoAP2 genes was down-regulated during flower development. These results suggest that DoAP2 genes may play roles in plant regeneration and flower development in D. officinale. Four DoAP2 genes (DoAP2-1 from euAP2, DoAP2-2 from euANT, and DoAP2-6 and DoAP2-11 from basal ANT) were selected for further analyses. The transcriptional activation of DoAP2-1, DoAP2-2, DoAP2-6 and DoAP2-11 proteins, which were localized in the nucleus of Arabidopsis thaliana mesophyll protoplasts, was further analyzed by a dual-luciferase reporter gene system in Nicotiana benthamiana leaves. Our data showed that pBD-DoAP2-1, pBD-DoAP2-2, pBD-DoAP2-6 and pBD-DoAP2-11 significantly repressed the expression of the LUC reporter compared with the negative control (pBD), suggesting that these DoAP2 proteins may act as transcriptional repressors in the nucleus of plant cells. Our findings on AP2 genes in D. officinale shed light on the function of AP2 genes in this orchid and other plant species.

Transcriptome and Metabolome Reveal Salt-Stress Responses of Leaf Tissues from Dendrobium officinale.

Dendrobium officinale Kimura et Migo is a precious traditional Chinese medicine. Despite D. officinale displaying a good salt-tolerance level, the yield and growth of D. officinale were impaired drastically by the increasing soil secondary salinization. The molecular mechanisms of D. officinale plants' adaptation to salt stress are not well documented. Therefore, in the present study, D. officinale plants were treated with 250 mM NaCl. Transcriptome analysis showed that salt stress significantly altered various metabolic pathways, including phenylalanine metabolism, flavonoid biosynthesis, and α-linolenic acid metabolism, and significantly upregulated the mRNA expression levels of DoAOC, DoAOS, DoLOX2S, DoMFP, and DoOPR involved in the jasmonic acid (JA) biosynthesis pathway, as well as rutin synthesis genes involved in the flavonoid synthesis pathway. In addition, metabolomics analysis showed that salt stress induced the accumulation of some compounds in D. officinale leaves, especially flavonoids, sugars, and alkaloids, which may play an important role in salt-stress responses of leaf tissues from D. officinale. Moreover, salt stress could trigger JA biosynthesis, and JA may act as a signal molecule that promotes flavonoid biosynthesis in D. officinale leaves. To sum up, D. officinale plants adapted to salt stress by enhancing the biosynthesis of secondary metabolites.

Comparative Transcriptomic and Metabolic Analyses Reveal the Molecular Mechanism of Ovule Development in the Orchid, Cymbidium sinense.

Ovule development is pivotal to plant reproduction and seed development. Cymbidium sinense (Orchidaceae) has high ornamental value due to its pleasant aroma and elegant floral morphology. The regulatory mechanism underlying ovule development in orchids, especially C. sinense, is largely unknown and information on the C. sinense genome is very scarce. In this study, a combined analysis was performed on the transcriptome and non-targeted metabolomes of 18 C. sinense 'Qi Jian Hei Mo' ovule samples. Transcriptome analysis assembled gene-related information related to six growth stages of C. sinense ovules (S1-S6, equivalent to 30, 35, 42, 46, 53, and 60 days after pollination). Illumina sequencing technology was used to obtain the complete set of transcriptome sequences of the 18 samples. A total of 81,585 unigene sequences were obtained after assembly, 24,860 (30.47%) of which were functionally annotated. Using transcriptome sequencing technology, a total of 9845 differentially expressed unigenes (DEUs) were identified in C. sinense ovules that were assigned to specific metabolic pathways according to the Kyoto Encyclopedia of Genes and Genomes (KEGG). DEUs associated with transcription factors (TFs) and phytohormones were identified and analyzed. The TFs homeobox and MADS-box were associated with C. sinense ovule development. In particular, the phytohormones associated with DEUs such as indole-3-acetic acid (IAA), cytokinin (CK), gibberellin (GA), abscisic acid (ABA), brassinosteroid (BR), and jasmonate (JA), may have important regulatory effects on C. sinense ovule development. Metabolomic analysis showed an inconsistent number of KEGG annotations of differential metabolites across comparisons (S2_vs_S4, S2_vs_S5, and S4_vs_S5 contained 23, 26, and 3 annotations, respectively) in C. sinense ovules. This study provides a valuable foundation for further understanding the regulation of orchid ovule development and formation, and establishes a theoretical background for future practical applications during orchid cultivation.

Functional Characterization of a Dendrobium officinale Geraniol Synthase DoGES1 Involved in Floral Scent Formation.

Floral scent is a key ornamental trait that determines the quality and commercial value of orchids. Geraniol, an important volatile monoterpene in orchids that attracts pollinators, is also involved in responses to stresses but the geraniol synthase (GES) responsible for its synthesis in the medicinal orchid Dendrobium officinale has not yet been identified. In this study, three potential geraniol synthases were mined from the D. officinale genome. DoGES1, which was localized in chloroplasts, was characterized as a geraniol synthase. DoGES1 was highly expressed in flowers, especially in petals. DoGES1 transcript levels were high in the budding stage of D. officinale flowers at 11:00 a.m. DoGES1 catalyzed geraniol in vitro, and transient expression of DoGES1 in Nicotiana benthamiana leaves resulted in the accumulation of geraniol in vivo. These findings on DoGES1 advance our understanding of geraniol biosynthesis in orchids, and lay the basis for genetic modification of floral scent in D. officinale or in other ornamental orchids.