ARGONAUTE10 controls cell fate specification and formative cell divisions in the Arabidopsis root.

A key question in plant biology is how oriented cell divisions are integrated with patterning mechanisms to generate organs with adequate cell type allocation. In the root vasculature, a gradient of miRNA165/6 controls the abundance of HD-ZIP III transcription factors, which in turn control cell fate and spatially restrict vascular cell proliferation to specific cells. Here, we show that vascular development requires the presence of ARGONAUTE10, which is thought to sequester miRNA165/6 and protect HD-ZIP III transcripts from degradation. Our results suggest that the miR165/6-AGO10-HDZIP III module acts by buffering cytokinin responses and restricting xylem differentiation. Mutants of AGO10 show faster growth rates and strongly enhanced survival under severe drought conditions. However, this superior performance is offset by markedly increased variation and phenotypic plasticity in sub-optimal carbon supply conditions. Thus, AGO10 is required for the control of formative cell division and coordination of robust cell fate specification of the vasculature, while altering its expression provides a means to adjust phenotypic plasticity.

Deep Learning-Based Automated Detection of Retinal Breaks and Detachments on Fundus Photography.

Purpose: The purpose of this study was to develop a deep learning algorithm, to detect retinal breaks and retinal detachments on ultra-widefield fundus (UWF) optos images using artificial intelligence (AI). Methods: Optomap UWF images of the database were annotated to four groups by two retina specialists: (1) retinal breaks without detachment, (2) retinal breaks with retinal detachment, (3) retinal detachment without visible retinal breaks, and (4) a combination of groups 1 to 3. The fundus image data set was split into a training set and an independent test set following an 80% to 20% ratio. Image preprocessing methods were applied. An EfficientNet classification model was trained with the training set and evaluated with the test set. Results: A total of 2489 UWF images were included into the dataset, resulting in a training set size of 2008 UWF images and a test set size of 481 images. The classification models achieved an area under the receiver operating characteristic curve (AUC) on the testing set of 0.975 regarding lesion detection, an AUC of 0.972 for retinal detachment and an AUC of 0.913 for retinal breaks. Conclusions: A deep learning system to detect retinal breaks and retinal detachment using UWF images is feasible and has a good specificity. This is relevant for clinical routine as there can be a high rate of missed breaks in clinics. Future clinical studies will be necessary to evaluate the cost-effectiveness of applying such an algorithm as an automated auxiliary tool in a large practices or tertiary referral centers. Translational Relevance: This study demonstrates the relevance of applying AI in diagnosing peripheral retinal breaks in clinical routine in UWF fundus images.

Cataract-1K Dataset for Deep-Learning-Assisted Analysis of Cataract Surgery Videos.

In recent years, the landscape of computer-assisted interventions and post-operative surgical video analysis has been dramatically reshaped by deep-learning techniques, resulting in significant advancements in surgeons' skills, operation room management, and overall surgical outcomes. However, the progression of deep-learning-powered surgical technologies is profoundly reliant on large-scale datasets and annotations. In particular, surgical scene understanding and phase recognition stand as pivotal pillars within the realm of computer-assisted surgery and post-operative assessment of cataract surgery videos. In this context, we present the largest cataract surgery video dataset that addresses diverse requisites for constructing computerized surgical workflow analysis and detecting post-operative irregularities in cataract surgery. We validate the quality of annotations by benchmarking the performance of several state-of-the-art neural network architectures for phase recognition and surgical scene segmentation. Besides, we initiate the research on domain adaptation for instrument segmentation in cataract surgery by evaluating cross-domain instrument segmentation performance in cataract surgery videos. The dataset and annotations are publicly available in Synapse.

A self-regulatory cell-wall-sensing module at cell edges controls plant growth.

Morphogenesis of multicellular organs requires coordination of cellular growth. In plants, cell growth is determined by turgor pressure and the mechanical properties of the cell wall, which also glues cells together. Because plants have to integrate tissue-scale mechanical stresses arising through growth in a fixed tissue topology, they need to monitor cell wall mechanical status and adapt growth accordingly. Molecular factors have been identified, but whether cell geometry contributes to wall sensing is unknown. Here we propose that plant cell edges act as cell-wall-sensing domains during growth. We describe two Receptor-Like Proteins, RLP4 and RLP4-L1, which occupy a unique polarity domain at cell edges established through a targeted secretory transport pathway. We show that RLP4s associate with the cell wall at edges via their extracellular domain, respond to changes in cell wall mechanics and contribute to directional growth control in Arabidopsis.

Epigenetic control over cell-intrinsic immune response antagonizes self-renewal in acute myeloid leukemia.

Epigenetic modulation of the cell-intrinsic immune response holds promise as a therapeutic approach for leukemia. However, current strategies designed for transcriptional activation of endogenous transposons and subsequent interferon type-I (IFN-I) response, show limited clinical efficacy. Histone lysine methylation is an epigenetic signature in IFN-I response associated with suppression of IFN-I and IFN stimulated genes, suggesting histone demethylation as key mechanism of reactivation. In this study, we unveil the histone demethylase PHF8 as a direct initiator and regulator of cell-intrinsic immune response in acute myeloid leukemia (AML). Site-specific phosphorylation of PHF8 orchestrates epigenetic changes that upregulate cytosolic RNA sensors, particularly the TRIM25-RIG-I-IFIT5 axis, thereby triggering the cellular IFN-I response-differentiation-apoptosis network. This signaling cascade largely counteracts differentiation block and growth of human AML cells across various disease subtypes in vitro and in vivo. Through proteome analysis of over 200 primary AML bone marrow samples, we identify a distinct PHF8/IFN-I signature in half of the patient population, without significant associations with known clinically or genetically defined AML subgroups. This profile was absent in healthy CD34-positive hematopoietic progenitor cells, suggesting therapeutic applicability in a large fraction of AML patients. Pharmacological support of PHF8 phosphorylation significantly impairs growth of primary AML patient samples. These findings provide novel opportunities for harnessing the cell-intrinsic immune response in the development of immunotherapeutic strategies against AML.

Identification and interpretation of gait analysis features and foot conditions by explainable AI.

Clinical gait analysis is a crucial step for identifying foot disorders and planning surgery. Automating this process is essential for efficiently assessing the substantial amount of gait data. In this study, we explored the potential of state-of-the-art machine learning (ML) and explainable artificial intelligence (XAI) algorithms to automate all various steps involved in gait analysis for six specific foot conditions. To address the complexity of gait data, we manually created new features, followed by recursive feature elimination using Support Vector Machines (SVM) and Random Forests (RF) to eliminate low-variance features. SVM, RF, K-nearest Neighbor (KNN), and Logistic Regression (LREGR) were compared for classification, with a Majority Voting (MV) model combining trained models. KNN and MV achieved mean balanced accuracy, recall, precision, and F1 score of 0.87. All models were interpreted using Local Interpretable Model-agnostic Explanation (LIME) method and the five most relevant features were identified for each foot condition. High success scores indicate a strong relationship between selected features and foot conditions, potentially indicating clinical relevance. The proposed ML pipeline, adaptable for other foot conditions, showcases its potential in aiding experts in foot condition identification and planning surgeries.

Comprehensive multi-gas study by means of fiber-enhanced Raman spectroscopy for the investigation of nitrogen cycle processes.

The extensive use of synthetic fertilizers has led to a considerable increase in reactive nitrogen input into agricultural and natural systems, resulting in negative effects in multiple ecosystems, the so-called nitrogen cascade. Since the global population relies on fertilization for food production, synthetic fertilizer use needs to be optimized by balancing crop yield and reactive nitrogen losses. Fiber-enhanced Raman spectroscopy (FERS) is introduced as a unique method for the simultaneous quantification of multiple gases to the study processes related to the nitrogen cycle. By monitoring changes in the headspace gas concentrations, processes such as denitrification, nitrification, respiration, and nitrogen fixation, as well as fertilizer addition were studied. The differences in concentration between the ambient and prepared process samples were evident in the Raman spectra, allowing for differentiation of process-specific spectra. Gas mixture concentrations were quantified within a range of low ppm to 100% for the gases N2, O2, CO2, N2O, and NH3. Compositional changes were attributed to processes of the nitrogen cycle. With help of multivariate curve resolution, it was possible to quantify N2O and CO2 simultaneously. The impact of fertilizers on N-cycle processes in soil was simulated and analyzed for identifying active processes. Thus, FERS was proven to be a suitable technique to optimize fertilizer composition and to quantify N2O and NH3 emissions, all with a single device and without further sample preparation.

Spectrally Resolved and Highly Parallelized Raman Difference Spectroscopy for the Analysis of Drug-Target Interactions between the Antimalarial Drug Chloroquine and Hematin.

Malaria is a severe disease caused by cytozoic parasites of the genus Plasmodium, which infiltrate and infect red blood cells. Several drugs have been developed to combat the devastating effects of malaria. Antimalarials based on quinolines inhibit the crystallization of hematin into hemozoin within the parasite, ultimately leading to its demise. Despite the frequent use of these agents, there are unanswered questions about their mechanisms of action. In the present study, the quinoline chloroquine and its interaction with the target structure hematin was investigated using an advanced, highly parallelized Raman difference spectroscopy (RDS) setup. Simultaneous recording of the spectra of hematin and chloroquine mixtures with varying compositions enabled the observation of changes in peak heights and positions based on the altered molecular structure resulting from their interaction. A shift of (-1.12 ± 0.05) cm-1 was observed in the core-size marker band ν(CαCm)asym peak position of the 1:1 chloroquine-hematin mixture compared to pure hematin. The oxidation-state marker band ν(pyrrole half-ring)sym exhibited a shift by (+0.93 ± 0.13) cm-1. These results were supported by density functional theory (DFT) calculations, indicating a hydrogen bond between the quinolinyl moiety of chloroquine and the oxygen atom of ferric protoporphyrin IX hydroxide (Fe(III)PPIX-OH). The consequence is a reduced electron density within the porphyrin moiety and an increase in its core size. This hypothesis provided further insights into the mechanism of hemozoin inhibition, suggesting chloroquine binding to the monomeric form of hematin, thereby preventing its further crystallization to hemozoin.

DeepPyramid+: medical image segmentation using Pyramid View Fusion and Deformable Pyramid Reception.

PURPOSE: Semantic segmentation plays a pivotal role in many applications related to medical image and video analysis. However, designing a neural network architecture for medical image and surgical video segmentation is challenging due to the diverse features of relevant classes, including heterogeneity, deformability, transparency, blunt boundaries, and various distortions. We propose a network architecture, DeepPyramid+, which addresses diverse challenges encountered in medical image and surgical video segmentation. METHODS: The proposed DeepPyramid+ incorporates two major modules, namely "Pyramid View Fusion" (PVF) and "Deformable Pyramid Reception" (DPR), to address the outlined challenges. PVF replicates a deduction process within the neural network, aligning with the human visual system, thereby enhancing the representation of relative information at each pixel position. Complementarily, DPR introduces shape- and scale-adaptive feature extraction techniques using dilated deformable convolutions, enhancing accuracy and robustness in handling heterogeneous classes and deformable shapes. RESULTS: Extensive experiments conducted on diverse datasets, including endometriosis videos, MRI images, OCT scans, and cataract and laparoscopy videos, demonstrate the effectiveness of DeepPyramid+ in handling various challenges such as shape and scale variation, reflection, and blur degradation. DeepPyramid+ demonstrates significant improvements in segmentation performance, achieving up to a 3.65% increase in Dice coefficient for intra-domain segmentation and up to a 17% increase in Dice coefficient for cross-domain segmentation. CONCLUSIONS: DeepPyramid+ consistently outperforms state-of-the-art networks across diverse modalities considering different backbone networks, showcasing its versatility. Accordingly, DeepPyramid+ emerges as a robust and effective solution, successfully overcoming the intricate challenges associated with relevant content segmentation in medical images and surgical videos. Its consistent performance and adaptability indicate its potential to enhance precision in computerized medical image and surgical video analysis applications.

Cytokine-responsive T- and NK-cells portray SARS-CoV-2 vaccine-responders and infection in multiple myeloma patients.

Patients with multiple myeloma (MM) routinely receive mRNA-based vaccines to reduce COVID-19-related mortality. However, whether disease- and therapy-related alterations in immune cells and cytokine-responsiveness contribute to the observed heterogeneous vaccination responses is unclear. Thus, we analyzed peripheral blood mononuclear cells from patients with MM during and after SARS-CoV-2 vaccination and breakthrough infection (BTI) using combined whole-transcriptome and surface proteome single-cell profiling with functional serological and T-cell validation in 58 MM patients. Our results demonstrate that vaccine-responders showed a significant overrepresentation of cytotoxic CD4+ T- and mature CD38+ NK-cells expressing FAS+/TIM3+ with a robust cytokine-responsiveness, such as type-I-interferon-, IL-12- and TNF-α-mediated signaling. Patients with MM experiencing BTI developed strong serological and cellular responses and exhibited similar cytokine-responsive immune cell patterns as vaccine-responders. This study can expand our understanding of molecular and cellular patterns associated with immunization responses and may benefit the design of improved vaccination strategies in immunocompromised patients.

Assessment of choriocapillaris and choroidal vascular changes in posterior uveitis using swept-source wide-field optical coherence tomography angiography.

PURPOSE: To evaluate choriocapillaris (CC) and choroidal vascular changes in patients with posterior uveitis using swept-source (SS) wide-field optical coherence tomography angiography (OCTA). METHOD: Consecutive patients with posterior uveitis were evaluated using 3×3 mm and 12×12 mm OCTA scan patterns and montage images of 5×12×12 mm or 2×15×9 mm, covering approximately 70°-90° of the retina. The images were quantitatively and qualitatively analysed and compared with healthy controls. RESULTS: Eighty-six eyes of 56 patients with posterior uveitis (mean age 45.2±19.9 years; 58.9% female), and 38 eyes of 19 age-matched healthy controls (57.9% female) were included. The mean perfusion density (PD) in 3×3 mm and 12×12 mm CC scans was significantly lower in eyes with posterior uveitis compared with those of healthy controls. However, no significant difference in the mean PD of choroidal scans was found comparing eyes with posterior uveitis and healthy controls. The mean PD in eyes with active disease was significantly higher compared with the inactive eyes on 12×12 mm choroidal scans (55.61% vs 51.25%, p=0.02), while no difference was found in the CC slabs. CONCLUSION: CC and choroidal assessment using OCTA provides useful information in patients with posterior uveitis. SS-OCTA metrics of the CC and choroidal slabs are promising tools in uveitis patients in the future. TRIAL REGISTRATION NUMBER: NCT02811536.

Deep learning predicts therapy-relevant genetics in acute myeloid leukemia from Pappenheim-stained bone marrow smears.

ABSTRACT: The detection of genetic aberrations is crucial for early therapy decisions in acute myeloid leukemia (AML) and recommended for all patients. Because genetic testing is expensive and time consuming, a need remains for cost-effective, fast, and broadly accessible tests to predict these aberrations in this aggressive malignancy. Here, we developed a novel fully automated end-to-end deep learning pipeline to predict genetic aberrations directly from single-cell images from scans of conventionally stained bone marrow smears already on the day of diagnosis. We used this pipeline to compile a multiterabyte data set of >2 000 000 single-cell images from diagnostic samples of 408 patients with AML. These images were then used to train convolutional neural networks for the prediction of various therapy-relevant genetic alterations. Moreover, we created a temporal test cohort data set of >444 000 single-cell images from further 71 patients with AML. We show that the models from our pipeline can significantly predict these subgroups with high areas under the curve of the receiver operating characteristic. Potential genotype-phenotype links were visualized with 2 different strategies. Our pipeline holds the potential to be used as a fast and inexpensive automated tool to screen patients with AML for therapy-relevant genetic aberrations directly from routine, conventionally stained bone marrow smears already on the day of diagnosis. It also creates a foundation to develop similar approaches for other bone marrow disorders in the future.

Functional knee joint axis calibration and outcome after femoral derotation in patients with cerebral palsy.

BACKGROUND: Patients with cerebral palsy and increased femoral anteversion frequently show disturbing internal rotation gait which may be treated via femoral derotation osteotomy (FDO). A recent study monitored that hip rotation in gait may heavily depend on the procedure by which it is being determined. Traditional measures using the femoral epicondyles as reference for the knee axis (CONV) resulted in more severe transverse plane deviations compared to those using a functional method (FUNC) with relevant implications for treatment indication of FDO. RESEARCH QUESTION: Is mean hip rotation in stance (mHipRotSt) as obtained via FUNC the more sensitive measure for explaining functional changes after FDO compared to CONV method taking the femoral epicondyles as reference for the knee axis? METHODS: 3D-gait analysis before and one year after FDO was performed in fourteen patients including functional joint axis determination of the knee of which MR imaging was available in eight patients both pre- and postoperatively. Transverse plane gait parameters were calculated using both approaches (CONV, FUNC). Differences between examinations as well as between methods were determined. RESULTS: Changes in femoral anteversion as measured by MR reasonably well confirm the structural changes as measured clinically and intraoperatively. The average change in mHipRotSt across the group was substantially smaller than the structural change implies. Further, using the FUNC approach led to much smaller values compared to when using the CONV approach. We address this to a mismatch between the axes determined in each method. SIGNIFICANCE: In the presence of femoral deformity, the knee joint axis as determined via a functional method together with the conventional method (femoral epicondyles for the knee axis) allows to quantify knee rotation independent of torsional parameters of the tibia. It may therefore help to better quantify rotational malalignments in gait and improve decision making of FDO.

The association of changes in leisure-time physical activity on depressive symptoms during Covid-19 in German adults: A longitudinal study.

This longitudinal survey assessed preventive and curative antidepressant effects of at least 75 min/week of leisure-time physical activity (LTPA) at different timepoints of the Covid-19 pandemic. We further investigated if high self-regulation skills were associated with higher odds of initiating or maintaining LTPA during the pandemic. Data was collected online from 4253 participants (age: m = 33.65 years, SD = 0.79; 79 % female) during the first quarantine measures in Germany (T1), as well as 4 weeks (T2), and 8 months (T3) later. We performed linear mixed models with changes (T2-T1, T3-T1) in LTPA and baseline major depressive disorder (MDD) as predictors (main effects and interaction effect) and depressive symptoms (at T2, T3) as the primary outcome. We found significant interaction effects of baseline depression and change in LTPA on depressive symptoms at T2 and T3 (p < 0.001). For probable cases of MDD an increasing LTPA to ≥75 min/week (vs. no change, <75 min/week) was associated with less depressive symptoms at T2 and T3 (p = 0.003, d = 0.28). For absence of depression at baseline, remaining at ≥75 min/week of LTPA was associated with less depressive symptoms at T2 and T3 compared to remaining at <75 min/week (p = 0.006, d = 0.11) or decreasing LTPA to <75 min/week (p = 0.018, d = 0.11). Reporting high self-regulation at T1 was associated with higher odds of performing ≥75 min/week of LTPA at T2/T3 (OR = 1.74, p < 0.001). In general, studies report reduced LTPA during Covid-19. To benefit from the reported preventive and interventional effects, further interventions should focus on improving physical activity related self-regulation to identify and overcome barriers for LTPA.

A new oxidative pathway of nitric oxide production from oximes in plants.

Nitric oxide (NO) is an essential reactive oxygen species and a signal molecule in plants. Although several studies have proposed the occurrence of oxidative NO production, only reductive routes for NO production, such as the nitrate (NO-3) -upper-reductase pathway, have been evidenced to date in land plants. However, plants grown axenically with ammonium as the sole source of nitrogen exhibit contents of nitrite and NO3-, evidencing the existence of a metabolic pathway for oxidative production of NO. We hypothesized that oximes, such as indole-3-acetaldoxime (IAOx), a precursor to indole-3-acetic acid, are intermediate oxidation products in NO synthesis. We detected the production of NO from IAOx and other oximes catalyzed by peroxidase (POD) enzyme using both 4-amino-5-methylamino-2',7'-difluorescein fluorescence and chemiluminescence. Flavins stimulated the reaction, while superoxide dismutase inhibited it. Interestingly, mouse NO synthase can also use IAOx to produce NO at a lower rate than POD. We provided a full mechanism for POD-dependent NO production from IAOx consistent with the experimental data and supported by density functional theory calculations. We showed that the addition of IAOx to extracts from Medicago truncatula increased the in vitro production of NO, while in vivo supplementation of IAOx and other oximes increased the number of lateral roots, as shown for NO donors, and a more than 10-fold increase in IAOx dehydratase expression. Furthermore, we found that in vivo supplementation of IAOx increased NO production in Arabidopsis thaliana wild-type plants, while prx33-34 mutant plants, defective in POD33-34, had reduced production. Our data show that the release of NO by IAOx, as well as its auxinic effect, explain the superroot phenotype. Collectively, our study reveals that plants produce NO utilizing diverse molecules such as oximes, POD, and flavins, which are widely distributed in the plant kingdom, thus introducing a long-awaited oxidative pathway to NO production in plants. This knowledge has essential implications for understanding signaling in biological systems.

Video aided extinction measurement: A competitive method for dust density diagnostics.

Among the most popular methods to measure dust density in a laboratory setup are 1D extinction, Abel inversion for circularly symmetric geometries, and computer tomography (CT) for arbitrary geometries. We present a new method based on a 1D extinction measurement in correlation with a video taken at an acute angle. It works well with limited optical access and has a good time resolution (at least several hertz). It measures the dust density within a slice of a nanodust cloud with precision comparable to other methods. Depending on the setup, this video aided extinction measurement can replace CT.

Influence of perioperative step volume on complication rate and length of hospital stay after colorectal cancer surgery (IPOS trial): study protocol for a randomised controlled single-centre trial at a German university hospital.

BACKGROUND: Perioperative mobilisation and physical activity are critical components of postoperative rehabilitation. Physical inactivity is a significant risk factor for complications and prolonged hospitalisation. However, specific recommendations for preoperative and postoperative physical activity levels are currently lacking. Evidence suggests that daily step count before and after surgery may impact the length of hospital stay and complication rate.The goal of this study is to determine the effectiveness of perioperative step volume recommendations, measured by pedometers, in reducing the length of hospital stay and complication rate for patients undergoing colorectal cancer surgery. METHODS: This study is a single-centre randomised controlled trial with two arms, allocated at a 1:1 ratio. The trial includes individuals undergoing colorectal surgery for either suspected or confirmed colorectal malignancy. A total of 222 patients will be randomly assigned to either an intervention or a control group. Step counts will be measured using a pedometer. Patients assigned to the intervention group will be given a predetermined preoperative and postoperative step count goal. The analysis will be conducted on preoperative and postoperative physical activity, quality of life, health, duration of hospitalisation, complication rate and bowel function, among other factors. ETHICS AND DISSEMINATION: The trial was approved by the ethics committee of the Ludwig-Maximilians-University of Munich, Germany (reference number: 22-0758, protocol version 2022.02). Results will be published in peer-reviewed journals and shared at academic conferences. After the publication of the results, a fully anonymised data set and the statistical code can be made available on justified scientific request and after ethical approval has been granted. TRIAL REGISTRATION NUMBER: DRKS00030017.

SH2 domain-containing inositol 5-phosphatases support the survival of Burkitt lymphoma cells by promoting energy metabolism.

Burkitt lymphoma cells (BL) exploit antigen-independent tonic signals transduced by the B cell antigen receptor (BCR) for their survival, but the molecular details of the rewired BLspecific BCR signal network remain unclear. A loss of function screen revealed the SH2 domain-containing 5`-inositol phosphatase 2 (SHIP2) as a potential modulator of BL fitness. We characterized the role of SHIP2 in BL survival in several BL cell models and show that perturbing SHIP2 function renders cells more susceptible to apoptosis, while attenuating proliferation in a BCR-dependent manner. Unexpectedly, SHIP2 deficiency did neither affect PI3K survival signals nor MAPK activity, but attenuated ATP production. We found that an efficient energy metabolism in BL cells requires phosphatidylinositol-3,4-bisphosphate (PI(3,4)P2), which is the enzymatic product of SHIP proteins. Consistently, interference with the function of SHIP1 and SHIP2 augments BL cell susceptibility to PI3K inhibition. Notably, we here provide a molecular basis of how tonic BCR signals are connected to energy supply, which is particularly important for such an aggressively growing neoplasia. These findings may help to improve therapies for the treatment of BL by limiting energy metabolism through the inhibition of SHIP proteins, which renders BL cells more susceptible to the targeting of survival signals.

Predicting OCT biological marker localization from weak annotations.

Recent developments in deep learning have shown success in accurately predicting the location of biological markers in Optical Coherence Tomography (OCT) volumes of patients with Age-Related Macular Degeneration (AMD) and Diabetic Retinopathy (DR). We propose a method that automatically locates biological markers to the Early Treatment Diabetic Retinopathy Study (ETDRS) rings, only requiring B-scan-level presence annotations. We trained a neural network using 22,723 OCT B-Scans of 460 eyes (433 patients) with AMD and DR, annotated with slice-level labels for Intraretinal Fluid (IRF) and Subretinal Fluid (SRF). The neural network outputs were mapped into the corresponding ETDRS rings. We incorporated the class annotations and domain knowledge into a loss function to constrain the output with biologically plausible solutions. The method was tested on a set of OCT volumes with 322 eyes (189 patients) with Diabetic Macular Edema, with slice-level SRF and IRF presence annotations for the ETDRS rings. Our method accurately predicted the presence of IRF and SRF in each ETDRS ring, outperforming previous baselines even in the most challenging scenarios. Our model was also successfully applied to en-face marker segmentation and showed consistency within C-scans, despite not incorporating volume information in the training process. We achieved a correlation coefficient of 0.946 for the prediction of the IRF area.

Increased photosynthesis during spring drought in energy-limited ecosystems.

Drought is often thought to reduce ecosystem photosynthesis. However, theory suggests there is potential for increased photosynthesis during meteorological drought, especially in energy-limited ecosystems. Here, we examine the response of photosynthesis (gross primary productivity, GPP) to meteorological drought across the water-energy limitation spectrum. We find a consistent increase in eddy covariance GPP during spring drought in energy-limited ecosystems (83% of the energy-limited sites). Half of spring GPP sensitivity to precipitation was predicted solely from the wetness index (R2 = 0.47, p < 0.001), with weaker relationships in summer and fall. Our results suggest GPP increases during spring drought for 55% of vegetated Northern Hemisphere lands ( >30° N). We then compare these results to terrestrial biosphere model outputs and remote sensing products. In contrast to trends detected in eddy covariance data, model mean GPP always declined under spring precipitation deficits after controlling for air temperature and light availability. While remote sensing products captured the observed negative spring GPP sensitivity in energy-limited ecosystems, terrestrial biosphere models proved insufficiently sensitive to spring precipitation deficits.