Are Macula or Optic Nerve Head Structures Better at Diagnosing Glaucoma? An Answer Using Artificial Intelligence and Wide-Field Optical Coherence Tomography.

Purpose: We wanted to develop a deep-learning algorithm to automatically segment optic nerve head (ONH) and macula structures in three-dimensional (3D) wide-field optical coherence tomography (OCT) scans and to assess whether 3D ONH or macula structures (or a combination of both) provide the best diagnostic power for glaucoma. Methods: A cross-sectional comparative study was performed using 319 OCT scans of glaucoma eyes and 298 scans of nonglaucoma eyes. Scans were compensated to improve deep-tissue visibility. We developed a deep-learning algorithm to automatically label major tissue structures, trained with 270 manually annotated B-scans. The performance was assessed using the Dice coefficient (DC). A glaucoma classification algorithm (3D-CNN) was then designed using 500 OCT volumes and corresponding automatically segmented labels. This algorithm was trained and tested on three datasets: cropped scans of macular tissues, those of ONH tissues, and wide-field scans. The classification performance for each dataset was reported using the area under the curve (AUC). Results: Our segmentation algorithm achieved a DC of 0.94 ± 0.003. The classification algorithm was best able to diagnose glaucoma using wide-field scans, followed by ONH scans, and finally macula scans, with AUCs of 0.99 ± 0.01, 0.93 ± 0.06 and 0.91 ± 0.11, respectively. Conclusions: This study showed that wide-field OCT may allow for significantly improved glaucoma diagnosis over typical OCTs of the ONH or macula. Translational Relevance: This could lead to mainstream clinical adoption of 3D wide-field OCT scan technology.

AI-based clinical assessment of optic nerve head robustness superseding biomechanical testing.

BACKGROUND/AIMS: To use artificial intelligence (AI) to: (1) exploit biomechanical knowledge of the optic nerve head (ONH) from a relatively large population; (2) assess ONH robustness (ie, sensitivity of the ONH to changes in intraocular pressure (IOP)) from a single optical coherence tomography (OCT) volume scan of the ONH without the need for biomechanical testing and (3) identify what critical three-dimensional (3D) structural features dictate ONH robustness. METHODS: 316 subjects had their ONHs imaged with OCT before and after acute IOP elevation through ophthalmo-dynamometry. IOP-induced lamina cribrosa (LC) deformations were then mapped in 3D and used to classify ONHs. Those with an average effective LC strain superior to 4% were considered fragile, while those with a strain inferior to 4% robust. Learning from these data, we compared three AI algorithms to predict ONH robustness strictly from a baseline (undeformed) OCT volume: (1) a random forest classifier; (2) an autoencoder and (3) a dynamic graph convolutional neural network (DGCNN). The latter algorithm also allowed us to identify what critical 3D structural features make a given ONH robust. RESULTS: All three methods were able to predict ONH robustness from a single OCT volume scan alone and without the need to perform biomechanical testing. The DGCNN (area under the curve (AUC): 0.76±0.08) outperformed the autoencoder (AUC: 0.72±0.09) and the random forest classifier (AUC: 0.69±0.05). Interestingly, to assess ONH robustness, the DGCNN mainly used information from the scleral canal and the LC insertion sites. CONCLUSIONS: We propose an AI-driven approach that can assess the robustness of a given ONH solely from a single OCT volume scan of the ONH, and without the need to perform biomechanical testing. Longitudinal studies should establish whether ONH robustness could help us identify fast visual field loss progressors. PRECIS: Using geometric deep learning, we can assess optic nerve head robustness (ie, sensitivity to a change in IOP) from a standard OCT scan that might help to identify fast visual field loss progressors.

Calibrating machine learning approaches for probability estimation: A comprehensive comparison.

Statistical prediction models have gained popularity in applied research. One challenge is the transfer of the prediction model to a different population which may be structurally different from the model for which it has been developed. An adaptation to the new population can be achieved by calibrating the model to the characteristics of the target population, for which numerous calibration techniques exist. In view of this diversity, we performed a systematic evaluation of various popular calibration approaches used by the statistical and the machine learning communities for estimating two-class probabilities. In this work, we first provide a review of the literature and, second, present the results of a comprehensive simulation study. The calibration approaches are compared with respect to their empirical properties and relationships, their ability to generalize precise probability estimates to external populations and their availability in terms of easy-to-use software implementations. Third, we provide code from real data analysis allowing its application by researchers. Logistic calibration and beta calibration, which estimate an intercept plus one and two slope parameters, respectively, consistently showed the best results in the simulation studies. Calibration on logit transformed probability estimates generally outperformed calibration methods on nontransformed estimates. In case of structural differences between training and validation data, re-estimation of the entire prediction model should be outweighted against sample size of the validation data. We recommend regression-based calibration approaches using transformed probability estimates, where at least one slope is estimated in addition to an intercept for updating probability estimates in validation studies.

scRNA-sequencing in chick suggests a probabilistic model for cell fate allocation at the neural plate border.

The vertebrate 'neural plate border' is a transient territory located at the edge of the neural plate containing precursors for all ectodermal derivatives: the neural plate, neural crest, placodes and epidermis. Elegant functional experiments in a range of vertebrate models have provided an in-depth understanding of gene regulatory interactions within the ectoderm. However, these experiments conducted at tissue level raise seemingly contradictory models for fate allocation of individual cells. Here, we carry out single cell RNA sequencing of chick ectoderm from primitive streak to neurulation stage, to explore cell state diversity and heterogeneity. We characterise the dynamics of gene modules, allowing us to model the order of molecular events which take place as ectodermal fates segregate. Furthermore, we find that genes previously classified as neural plate border 'specifiers' typically exhibit dynamic expression patterns and are enriched in either neural, neural crest or placodal fates, revealing that the neural plate border should be seen as a heterogeneous ectodermal territory and not a discrete transitional transcriptional state. Analysis of neural, neural crest and placodal markers reveals that individual NPB cells co-express competing transcriptional programmes suggesting that their ultimate identify is not yet fixed. This population of 'border located undecided progenitors' (BLUPs) gradually diminishes as cell fate decisions take place. Considering our findings, we propose a probabilistic model for cell fate choice at the neural plate border. Our data suggest that the probability of a progenitor's daughters to contribute to a given ectodermal derivative is related to the balance of competing transcriptional programmes, which in turn are regulated by the spatiotemporal position of a progenitor.

Three-Dimensional Structural Phenotype of the Optic Nerve Head as a Function of Glaucoma Severity.

Importance: The 3-dimensional (3-D) structural phenotype of glaucoma as a function of severity was thoroughly described and analyzed, enhancing understanding of its intricate pathology beyond current clinical knowledge. Objective: To describe the 3-D structural differences in both connective and neural tissues of the optic nerve head (ONH) between different glaucoma stages using traditional and artificial intelligence-driven approaches. Design, Setting, and Participants: This cross-sectional, clinic-based study recruited 541 Chinese individuals receiving standard clinical care at Singapore National Eye Centre, Singapore, and 112 White participants of a prospective observational study at Vilnius University Hospital Santaros Klinikos, Vilnius, Lithuania. The study was conducted from May 2022 to January 2023. All participants had their ONH imaged using spectral-domain optical coherence tomography and had their visual field assessed by standard automated perimetry. Main Outcomes and Measures: (1) Clinician-defined 3-D structural parameters of the ONH and (2) 3-D structural landmarks identified by geometric deep learning that differentiated ONHs among 4 groups: no glaucoma, mild glaucoma (mean deviation [MD], ≥-6.00 dB), moderate glaucoma (MD, -6.01 to -12.00 dB), and advanced glaucoma (MD, <-12.00 dB). Results: Study participants included 213 individuals without glaucoma (mean age, 63.4 years; 95% CI, 62.5-64.3 years; 126 females [59.2%]; 213 Chinese [100%] and 0 White individuals), 204 with mild glaucoma (mean age, 66.9 years; 95% CI, 66.0-67.8 years; 91 females [44.6%]; 178 Chinese [87.3%] and 26 White [12.7%] individuals), 118 with moderate glaucoma (mean age, 68.1 years; 95% CI, 66.8-69.4 years; 49 females [41.5%]; 97 Chinese [82.2%] and 21 White [17.8%] individuals), and 118 with advanced glaucoma (mean age, 68.5 years; 95% CI, 67.1-69.9 years; 43 females [36.4%]; 53 Chinese [44.9%] and 65 White [55.1%] individuals). The majority of ONH structural differences occurred in the early glaucoma stage, followed by a plateau effect in the later stages. Using a deep neural network, 3-D ONH structural differences were found to be present in both neural and connective tissues. Specifically, a mean of 57.4% (95% CI, 54.9%-59.9%, for no to mild glaucoma), 38.7% (95% CI, 36.9%-40.5%, for mild to moderate glaucoma), and 53.1 (95% CI, 50.8%-55.4%, for moderate to advanced glaucoma) of ONH landmarks that showed major structural differences were located in neural tissues with the remaining located in connective tissues. Conclusions and Relevance: This study uncovered complex 3-D structural differences of the ONH in both neural and connective tissues as a function of glaucoma severity. Future longitudinal studies should seek to establish a connection between specific 3-D ONH structural changes and fast visual field deterioration and aim to improve the early detection of patients with rapid visual field loss in routine clinical care.

Cost-effectiveness of a telemonitoring programme in patients with cardiovascular diseases compared with standard of care.

OBJECTIVES: The main aim of this work was to analyse the cost-effectiveness of an integrated care concept (NICC) that combines telemonitoring with the support of a care centre in addition to guideline therapy for patients. Secondary aims were to compare health utility and health-related quality of life (QoL) between NICC and standard of care (SoC). METHODS: The randomised controlled CardioCare MV Trial compared NICC and SoC in patients from Mecklenburg-West Pomerania (Germany) with atrial fibrillation, heart failure or treatment-resistant hypertension. QoL was measured using the EQ-5D-5L at baseline, 6 months and 1 year follow-up. Quality-adjusted life years (QALYs), EQ5D utility scores, Visual Analogue Scale (VAS) Scores and VAS adjusted life years (VAS-AL) were calculated. Cost data were obtained from health insurance companies, and the payer perspective was taken in health economic analyses. Quantile regression was used with adjustments for stratification variables. RESULTS: The net benefit of NICC (QALY) was 0.031 (95% CI 0.012 to 0.050; p=0.001) in this trial involving 957 patients. EQ5D Index values, VAS-ALs and VAS were larger for NICC compared with SoC at 1 year follow-up (all p≤0.004). Direct cost per patient and year were €323 (CI €157 to €489) lower in the NICC group. When 2000 patients are served by the care centre, NICC is cost-effective if one is willing to pay €10 652 per QALY per year. CONCLUSION: NICC was associated with higher QoL and health utility. The programme is cost-effective if one is willing to pay approximately €11 000 per QALY per year.

A gene regulatory network for neural induction.

During early vertebrate development, signals from a special region of the embryo, the organizer, can redirect the fate of non-neural ectoderm cells to form a complete, patterned nervous system. This is called neural induction and has generally been imagined as a single signalling event, causing a switch of fate. Here, we undertake a comprehensive analysis, in very fine time course, of the events following exposure of competent ectoderm of the chick to the organizer (the tip of the primitive streak, Hensen's node). Using transcriptomics and epigenomics we generate a gene regulatory network comprising 175 transcriptional regulators and 5614 predicted interactions between them, with fine temporal dynamics from initial exposure to the signals to expression of mature neural plate markers. Using in situ hybridization, single-cell RNA-sequencing, and reporter assays, we show that the gene regulatory hierarchy of responses to a grafted organizer closely resembles the events of normal neural plate development. The study is accompanied by an extensive resource, including information about conservation of the predicted enhancers in other vertebrates.

Medical Application of Geometric Deep Learning for the Diagnosis of Glaucoma.

Purpose: (1) To assess the performance of geometric deep learning in diagnosing glaucoma from a single optical coherence tomography (OCT) scan of the optic nerve head and (2) to compare its performance to that obtained with a three-dimensional (3D) convolutional neural network (CNN), and with a gold-standard parameter, namely, the retinal nerve fiber layer (RNFL) thickness. Methods: Scans of the optic nerve head were acquired with OCT for 477 glaucoma and 2296 nonglaucoma subjects. All volumes were automatically segmented using deep learning to identify seven major neural and connective tissues. Each optic nerve head was then represented as a 3D point cloud with approximately 1000 points. Geometric deep learning (PointNet) was then used to provide a glaucoma diagnosis from a single 3D point cloud. The performance of our approach (reported using the area under the curve [AUC]) was compared with that obtained with a 3D CNN, and with the RNFL thickness. Results: PointNet was able to provide a robust glaucoma diagnosis solely from a 3D point cloud (AUC = 0.95 ± 0.01).The performance of PointNet was superior to that obtained with a 3D CNN (AUC = 0.87 ± 0.02 [raw OCT images] and 0.91 ± 0.02 [segmented OCT images]) and with that obtained from RNFL thickness alone (AUC = 0.80 ± 0.03). Conclusions: We provide a proof of principle for the application of geometric deep learning in glaucoma. Our technique requires significantly less information as input to perform better than a 3D CNN, and with an AUC superior to that obtained from RNFL thickness. Translational Relevance: Geometric deep learning may help us to improve and simplify diagnosis and prognosis applications in glaucoma.

A decision-support tool for funding health innovations at a tertiary academic medical center.

OBJECTIVES: To report the processes used to design and implement an assessment tool to inform funding decisions for competing health innovations in a tertiary hospital. METHODS: We designed an assessment tool for health innovation proposals with three components: "value to the institution," "novelty," and "potential for adoption and scaling." The "value to the institution" component consisted of twelve weighted value attributes identified from the host institution's annual report; weights were allocated based on a survey of the hospital's leaders. The second and third components consisted of open-ended questions on "novelty" and "barriers to implementation" to support further dialogue. Purposive literature review was performed independently by two researchers for each assessment. The assessment tool was piloted during an institutional health innovation funding cycle. RESULTS: We used 17 days to evaluate ten proposals. The completed assessments were shared with an independent group of panellists, who selected five projects for funding. Proposals with the lowest scores for "value to the institution" had less perceived impact on the patient-related value attributes of "access," "patient centeredness," "health outcomes," "prevention," and "safety." Similar innovations were reported in literature in seven proposals; potential barriers to implementation were identified in six proposals. We included a worked example to illustrate the assessment process. CONCLUSIONS: We developed an assessment tool that is aligned with local institutional priorities. Our tool can augment the decision-making process when funding health innovation projects. The tool can be adapted by others facing similar challenges of trying to choose the best health innovations to fund.

Comparison of telemonitoring combined with intensive patient support with standard care in patients with chronic cardiovascular disease - a randomized clinical trial.

IMPORTANCE: Healthcare concepts for chronic diseases based on tele-monitoring have become increasingly important during COVID-19 pandemic. OBJECTIVE: To study the effectiveness of a novel integrated care concept (NICC) that combines tele-monitoring with the support of a call centre in addition to guideline therapy for patients with atrial fibrillation, heart failure, or treatment-resistant hypertension. DESIGN: A prospective, parallel-group, open-label, randomized, controlled trial. SETTING: Between December 2017 and August 2019 at the Rostock University Medical Center (Germany). PARTICIPANTS: Including 960 patients with either atrial fibrillation, heart failure, or treatment-resistant hypertension. INTERVENTIONS: Patients were randomized to either NICC (n = 478) or standard-of-care (SoC) (n = 482) in a 1:1 ratio. Patients in the NICC group received a combination of tele-monitoring and intensive follow-up and care through a call centre. MAIN OUTCOMES AND MEASURES: Three primary endpoints were formulated: (1) composite of all-cause mortality, stroke, and myocardial infarction; (2) number of inpatient days; (3) the first plus cardiac decompensation, all measured at 12-months follow-up. Superiority was evaluated using a hierarchical multiple testing strategy for the 3 primary endpoints, where the first step is to test the second primary endpoint (hospitalization) at two-sided 5%-significance level. In case of a non-significant difference between the groups for the rate of hospitalization, the superiority of NICC over SoC is not shown. RESULTS: The first primary endpoint occurred in 1.5% of NICC and 5.2% of SoC patients (OR: 3.3 [95%CI 1.4-8.3], p = 0.009). The number of inpatient treatment days did not differ significantly between both groups (p = 0.122). The third primary endpoint occurred in 3.6% of NICC and 8.1% of SoC patients (OR: 2.2 [95%CI 1.2-4.2], p = 0.016). Four patients died of all-cause death in the NICC and 23 in the SoC groups (OR: 4.4 [95%CI 1.6-12.6], p = 0.006). Based on the prespecified hierarchical statistical analysis protocol for multiple testing, the trial did not meet its primary outcome measure. CONCLUSIONS AND RELEVANCE: Among patients with atrial fibrillation, heart failure, or treatment-resistant hypertension, the NICC approach was not superior over SoC, despite a significant reduction in all-cause mortality, stroke, myocardial infarction and cardiac decompensation. Trial registration ClinicalTrials.gov Identifier: NCT03317951.

Geometric Deep Learning to Identify the Critical 3D Structural Features of the Optic Nerve Head for Glaucoma Diagnosis.

PURPOSE: To compare the performance of 2 relatively recent geometric deep learning techniques in diagnosing glaucoma from a single optical coherence tomographic (OCT) scan of the optic nerve head (ONH); and to identify the 3-dimensional (3D) structural features of the ONH that are critical for the diagnosis of glaucoma. DESIGN: Comparison and evaluation of deep learning diagnostic algorithms. METHODS: In this study, we included a total of 2247 nonglaucoma and 2259 glaucoma scans from 1725 participants. All participants had their ONHs imaged in 3D with Spectralis OCT. All OCT scans were automatically segmented using deep learning to identify major neural and connective tissues. Each ONH was then represented as a 3D point cloud. We used PointNet and dynamic graph convolutional neural network (DGCNN) to diagnose glaucoma from such 3D ONH point clouds and to identify the critical 3D structural features of the ONH for glaucoma diagnosis. RESULTS: Both the DGCNN (area under the curve [AUC]: 0.97±0.01) and PointNet (AUC: 0.95±0.02) were able to accurately detect glaucoma from 3D ONH point clouds. The critical points (ie, critical structural features of the ONH) formed an hourglass pattern, with most of them located within the neuroretinal rim in the inferior and superior quadrant of the ONH. CONCLUSIONS: The diagnostic accuracy of both geometric deep learning approaches was excellent. Moreover, we were able to identify the critical 3D structural features of the ONH for glaucoma diagnosis that tremendously improved the transparency and interpretability of our method. Consequently, our approach may have strong potential to be used in clinical applications for the diagnosis and prognosis of a wide range of ophthalmic disorders.

Discriminating Between Papilledema and Optic Disc Drusen Using 3D Structural Analysis of the Optic Nerve Head.

BACKGROUND AND OBJECTIVES: The distinction of papilledema from other optic nerve head (ONH) lesions mimicking papilledema, such as optic disc drusen (ODD), can be difficult in clinical practice. We aimed the following: (1) to develop a deep learning algorithm to automatically identify major structures of the ONH in 3-dimensional (3D) optical coherence tomography (OCT) scans and (2) to exploit such information to robustly differentiate among ODD, papilledema, and healthy ONHs. METHODS: This was a cross-sectional comparative study of patients from 3 sites (Singapore, Denmark, and Australia) with confirmed ODD, those with papilledema due to raised intracranial pressure, and healthy controls. Raster scans of the ONH were acquired using OCT imaging and then processed to improve deep-tissue visibility. First, a deep learning algorithm was developed to identify major ONH tissues and ODD regions. The performance of our algorithm was assessed using the Dice coefficient. Second, a classification algorithm (random forest) was designed to perform 3-class classifications (1: ODD, 2: papilledema, and 3: healthy ONHs) strictly from their drusen and prelamina swelling scores (calculated from the segmentations). To assess performance, we reported the area under the receiver operating characteristic curve for each class. RESULTS: A total of 241 patients (256 imaged ONHs, including 105 ODD, 51 papilledema, and 100 healthy ONHs) were retrospectively included in this study. Using OCT images of the ONH, our segmentation algorithm was able to isolate neural and connective tissues and ODD regions/conglomerates whenever present. This was confirmed by an averaged Dice coefficient of 0.93 ± 0.03 on the test set, corresponding to good segmentation performance. Classification was achieved with high AUCs, that is, 0.99 ± 0.001 for the detection of ODD, 0.99 ± 0.005 for the detection of papilledema, and 0.98 ± 0.01 for the detection of healthy ONHs. DISCUSSION: Our artificial intelligence approach can discriminate ODD from papilledema, strictly using a single OCT scan of the ONH. Our classification performance was very good in the studied population, with the caveat that validation in a much larger population is warranted. Our approach may have the potential to establish OCT imaging as one of the mainstays of diagnostic imaging for ONH disorders in neuro-ophthalmology, in addition to fundus photography.

Comparison of calling pipelines for whole genome sequencing: an empirical study demonstrating the importance of mapping and alignment.

Rapid advances in high-throughput DNA sequencing technologies have enabled the conduct of whole genome sequencing (WGS) studies, and several bioinformatics pipelines have become available. The aim of this study was the comparison of 6 WGS data pre-processing pipelines, involving two mapping and alignment approaches (GATK utilizing BWA-MEM2 2.2.1, and DRAGEN 3.8.4) and three variant calling pipelines (GATK 4.2.4.1, DRAGEN 3.8.4 and DeepVariant 1.1.0). We sequenced one genome in a bottle (GIAB) sample 70 times in different runs, and one GIAB trio in triplicate. The truth set of the GIABs was used for comparison, and performance was assessed by computation time, F1 score, precision, and recall. In the mapping and alignment step, the DRAGEN pipeline was faster than the GATK with BWA-MEM2 pipeline. DRAGEN showed systematically higher F1 score, precision, and recall values than GATK for single nucleotide variations (SNVs) and Indels in simple-to-map, complex-to-map, coding and non-coding regions. In the variant calling step, DRAGEN was fastest. In terms of accuracy, DRAGEN and DeepVariant performed similarly and both superior to GATK, with slight advantages for DRAGEN for Indels and for DeepVariant for SNVs. The DRAGEN pipeline showed the lowest Mendelian inheritance error fraction for the GIAB trios. Mapping and alignment played a key role in variant calling of WGS, with the DRAGEN outperforming GATK.

Detection of m6A from direct RNA sequencing using a multiple instance learning framework.

RNA modifications such as m6A methylation form an additional layer of complexity in the transcriptome. Nanopore direct RNA sequencing can capture this information in the raw current signal for each RNA molecule, enabling the detection of RNA modifications using supervised machine learning. However, experimental approaches provide only site-level training data, whereas the modification status for each single RNA molecule is missing. Here we present m6Anet, a neural-network-based method that leverages the multiple instance learning framework to specifically handle missing read-level modification labels in site-level training data. m6Anet outperforms existing computational methods, shows similar accuracy as experimental approaches, and generalizes with high accuracy to different cell lines and species without retraining model parameters. In addition, we demonstrate that m6Anet captures the underlying read-level stoichiometry, which can be used to approximate differences in modification rates. Overall, m6Anet offers a tool to capture the transcriptome-wide identification and quantification of m6A from a single run of direct RNA sequencing.

Sox8 remodels the cranial ectoderm to generate the ear.

The vertebrate inner ear arises from a pool of progenitors with the potential to contribute to all the sense organs and cranial ganglia in the head. Here, we explore the molecular mechanisms that control ear specification from these precursors. Using a multiomics approach combined with loss-of-function experiments, we identify a core transcriptional circuit that imparts ear identity, along with a genome-wide characterization of noncoding elements that integrate this information. This analysis places the transcription factor Sox8 at the top of the ear determination network. Introducing Sox8 into the cranial ectoderm not only converts non-ear cells into ear progenitors but also activates the cellular programs for ear morphogenesis and neurogenesis. Thus, Sox8 has the unique ability to remodel transcriptional networks in the cranial ectoderm toward ear identity.

The genomic origins of the world's first farmers.

The precise genetic origins of the first Neolithic farming populations in Europe and Southwest Asia, as well as the processes and the timing of their differentiation, remain largely unknown. Demogenomic modeling of high-quality ancient genomes reveals that the early farmers of Anatolia and Europe emerged from a multiphase mixing of a Southwest Asian population with a strongly bottlenecked western hunter-gatherer population after the last glacial maximum. Moreover, the ancestors of the first farmers of Europe and Anatolia went through a period of extreme genetic drift during their westward range expansion, contributing highly to their genetic distinctiveness. This modeling elucidates the demographic processes at the root of the Neolithic transition and leads to a spatial interpretation of the population history of Southwest Asia and Europe during the late Pleistocene and early Holocene.

An epithelial signalling centre in sharks supports homology of tooth morphogenesis in vertebrates.

Development of tooth shape is regulated by the enamel knot signalling centre, at least in mammals. Fgf signalling regulates differential proliferation between the enamel knot and adjacent dental epithelia during tooth development, leading to formation of the dental cusp. The presence of an enamel knot in non-mammalian vertebrates is debated given differences in signalling. Here, we show the conservation and restriction of fgf3, fgf10, and shh to the sites of future dental cusps in the shark (Scyliorhinus canicula), whilst also highlighting striking differences between the shark and mouse. We reveal shifts in tooth size, shape, and cusp number following small molecule perturbations of canonical Wnt signalling. Resulting tooth phenotypes mirror observed effects in mammals, where canonical Wnt has been implicated as an upstream regulator of enamel knot signalling. In silico modelling of shark dental morphogenesis demonstrates how subtle changes in activatory and inhibitory signals can alter tooth shape, resembling developmental phenotypes and cusp shapes observed following experimental Wnt perturbation. Our results support the functional conservation of an enamel knot-like signalling centre throughout vertebrates and suggest that varied tooth types from sharks to mammals follow a similar developmental bauplan. Lineage-specific differences in signalling are not sufficient in refuting homology of this signalling centre, which is likely older than teeth themselves.

The three-dimensional structural configuration of the central retinal vessel trunk and branches as a glaucoma biomarker.

PURPOSE: To assess whether the 3-dimensional (3D) structural configuration of the central retinal vessel trunk and its branches (CRVT&B) could be used as a diagnostic marker for glaucoma. DESIGN: Retrospective, deep-learning approach diagnosis study. METHODS: We trained a deep learning network to automatically segment the CRVT&B from the B-scans of the optical coherence tomography (OCT) volume of the optic nerve head. Subsequently, 2 different approaches were used for glaucoma diagnosis using the structural configuration of the CRVT&B as extracted from the OCT volumes. In the first approach, we aimed to provide a diagnosis using only 3D convolutional neural networks and the 3D structure of the CRVT&B. For the second approach, we projected the 3D structure of the CRVT&B orthographically onto sagittal, frontal, and transverse planes to obtain 3 two-dimensional (2D) images, and then a 2D convolutional neural network was used for diagnosis. The segmentation accuracy was evaluated using the Dice coefficient, whereas the diagnostic accuracy was assessed using the area under the receiver operating characteristic curves (AUCs). The diagnostic performance of the CRVT&B was also compared with that of retinal nerve fiber layer (RNFL) thickness (calculated in the same cohorts). RESULTS: Our segmentation network was able to efficiently segment retinal blood vessels from OCT scans. On a test set, we achieved a Dice coefficient of 0.81 ± 0.07. The 3D and 2D diagnostic networks were able to differentiate glaucoma from nonglaucoma subjects with accuracies of 82.7% and 83.3%, respectively. The corresponding AUCs for the CRVT&B were 0.89 and 0.90, higher than those obtained with RNFL thickness alone (AUCs ranging from 0.74 to 0.80). CONCLUSIONS: Our work demonstrated that the diagnostic power of the CRVT&B is superior to that of a gold-standard glaucoma parameter, that is, RNFL thickness. Our work also suggested that the major retinal blood vessels form a "skeleton"-the configuration of which may be representative of major optic nerve head structural changes as typically observed with the development and progression of glaucoma.

Apathy is associated with white matter network disruption and specific cognitive deficits in Parkinson's disease.

BACKGROUND: Apathy is common in Parkinson's disease (PD) but its underlying white matter (WM) architecture is not well understood. Moreover, how apathy affects cognitive functions in PD remains unclear. We investigated apathy-related WM network alterations and the impact of apathy on cognition in the context of PD. METHODS: Apathetic PD patients (aPD), non-apathetic PD patients (naPD), and matched healthy controls (HCs) underwent brain scans and clinical assessment. Graph-theoretical and network-based analyses were used for group comparisons of WM features derived from diffusion spectrum imaging (DSI). Path analysis was used to determine the direct and indirect effects of apathy and other correlates on different cognitive functions. RESULTS: The aPD group was impaired on neural integration measured by global efficiency (p = 0.009) and characteristic path length (p = 0.04), executive function (p < 0.001), episodic memory (p < 0.001) and visuospatial ability (p = 0.02), and had reduced connectivity between the bilateral parietal lobes and between the putamen and temporal regions (p < 0.05). In PD, executive function was directly impacted by apathy and motor severity and indirectly influenced by depression; episodic memory was directly and indirectly impacted by apathy and depression, respectively; conversely, visuospatial ability was not related to any of these factors. Neural integration, though being marginally correlated with apathy, was not associated with cognition. CONCLUSIONS: Our results suggest compromised neural integration and reduced structural connectivity in aPD. Apathy, depression, and motor severity showed distinct impacts on different cognitive functions with apathy being the most influential determinant of cognition in PD.

Describing the Structural Phenotype of the Glaucomatous Optic Nerve Head Using Artificial Intelligence.

PURPOSE: To develop a novel deep-learning approach that can describe the structural phenotype of the glaucomatous optic nerve head (ONH) and can be used as a robust glaucoma diagnosis tool. DESIGN: Retrospective, deep-learning approach diagnosis study. METHOD: We trained a deep-learning network to segment 3 neural-tissue and 4 connective-tissue layers of the ONH. The segmented optical coherence tomography images were then processed by a customized autoencoder network with an additional parallel branch for binary classification. The encoder part of the autoencoder reduced the segmented optical coherence tomography images into a low-dimensional latent space (LS), whereas the decoder and the classification branches reconstructed the images and classified them as glaucoma or nonglaucoma, respectively. We performed principal component analysis on the latent parameters and identified the principal components (PCs). Subsequently, the magnitude of each PC was altered in steps and reported how it impacted the morphology of the ONH. RESULTS: The image reconstruction quality and diagnostic accuracy increased with the size of the LS. With 54 parameters in the LS, the diagnostic accuracy was 92.0 ± 2.3% with a sensitivity of 90.0 ± 2.4% (at 95% specificity), and the corresponding Dice coefficient for the reconstructed images was 0.86 ± 0.04. By changing the magnitudes of PC in steps, we were able to reveal how the morphology of the ONH changes as one transitions from a "nonglaucoma" to a "glaucoma" condition. CONCLUSIONS: Our network was able to identify novel biomarkers of the ONH for glaucoma diagnosis. Specifically, the structural features identified by our algorithm were found to be related to clinical observations of glaucoma.