Using a new artificial intelligence-aided method to assess body composition CT segmentation in colorectal cancer patients.

INTRODUCTION: This study aimed to evaluate the accuracy of our own artificial intelligence (AI)-generated model to assess automated segmentation and quantification of body composition-derived computed tomography (CT) slices from the lumber (L3) region in colorectal cancer (CRC) patients. METHODS: A total of 541 axial CT slices at the L3 vertebra were retrospectively collected from 319 patients with CRC diagnosed during 2012-2019 at a single Australian tertiary institution, Western Health in Melbourne. A two-dimensional U-Net convolutional network was trained on 338 slices to segment muscle, visceral adipose tissue (VAT) and subcutaneous adipose tissue (SAT). Manual reading of these same slices of muscle, VAT and SAT was created to serve as ground truth data. The Dice similarity coefficient was used to assess the U-Net-based segmentation performance on both a validation dataset (68 slices) and a test dataset (203 slices). The measurement of cross-sectional area and Hounsfield unit (HU) density of muscle, VAT and SAT were compared between two methods. RESULTS: The segmentation for muscle, VAT and SAT demonstrated excellent performance for both the validation (Dice similarity coefficients >0.98, respectively) and test (Dice similarity coefficients >0.97, respectively) datasets. There was a strong positive correlation between manual and AI segmentation measurements of body composition for both datasets (Spearman's correlation coefficients: 0.944-0.999, P < 0.001). CONCLUSIONS: Compared to the gold standard, this fully automated segmentation system exhibited a high accuracy for assessing segmentation and quantification of abdominal muscle and adipose tissues of CT slices at the L3 in CRC patients.

Body composition assessment by artificial intelligence can be a predictive tool for short-term postoperative complications in Hartmann's reversals.

BACKGROUND: Hartmann's reversal, a complex elective surgery, reverses and closes the colostomy in individuals who previously underwent a Hartmann's procedure due to colonic pathology like cancer or diverticulitis. It demands careful planning and patient optimisation to help reduce postoperative complications. Preoperative evaluation of body composition has been useful in identifying patients at high risk of short-term postoperative outcomes following colorectal cancer surgery. We sought to explore the use of our in-house derived Artificial Intelligence (AI) algorithm to measure body composition within patients undergoing Hartmann's reversal procedure in the prediction of short-term postoperative complications. METHODS: A retrospective study of all patients who underwent Hartmann's reversal within a single tertiary referral centre (Western) in Melbourne, Australia and who had a preoperative Computerised Tomography (CT) scan performed. Body composition was measured using our previously validated AI algorithm for body segmentation developed by the Department of Surgery, Western Precinct, University of Melbourne. Sarcopenia in our study was defined as a skeletal muscle index (SMI), calculated as Skeletal Muscle Area (SMA) /height2 < 38.5 cm2/m2 in women and < 52.4 cm2/m2 in men. RESULTS: Between 2010 and 2020, 47 patients (mean age 63.1 ± 12.3 years; male, n = 28 (59.6%) underwent body composition analysis. Twenty-one patients (44.7%) were sarcopenic, and 12 (25.5%) had evidence of sarcopenic obesity. The most common postoperative complication was surgical site infection (SSI) (n = 8, 17%). Sarcopenia (n = 7, 87.5%, p = 0.02) and sarcopenic obesity (n = 5, 62.5%, p = 0.02) were significantly associated with SSIs. The risks of developing an SSI were 8.7 times greater when sarcopenia was present. CONCLUSION: Sarcopenia and sarcopenic obesity were related to postoperative complications following Hartmann's reversal. Body composition measured by a validated AI algorithm may be a beneficial tool for predicting short-term surgical outcomes for these patients.

Artificial intelligence measured 3D body composition to predict pathological response in rectal cancer patients.

BACKGROUND: The treatment of locally advanced rectal cancer (LARC) is moving towards total neoadjuvant therapy and potential organ preservation. Of particular interest are predictors of pathological complete response (pCR) that can guide personalized treatment. There are currently no clinical biomarkers which can accurately predict neoadjuvant therapy (NAT) response but body composition (BC) measures present as an emerging contender. The primary aim of the study was to determine if artificial intelligence (AI) derived body composition variables can predict pCR in patients with LARC. METHODS: LARC patients who underwent NAT followed by surgery from 2012 to 2023 were identified from the Australian Comprehensive Cancer Outcomes and Research Database registry (ACCORD). A validated in-house pre-trained 3D AI model was used to measure body composition via computed tomography images of the entire Lumbar-3 vertebral level to produce a volumetric measurement of visceral fat (VF), subcutaneous fat (SCF) and skeletal muscle (SM). Multivariate analysis between patient body composition and histological outcomes was performed. RESULTS: Of 214 LARC patients treated with NAT, 22.4% of patients achieved pCR. SM volume (P = 0.015) and age (P = 0.03) were positively associated with pCR in both male and female patients. SCF volume was associated with decreased likelihood of pCR (P = 0.059). CONCLUSION: This is the first study in the literature utilizing AI-measured 3D Body composition in LARC patients to assess their impact on pathological response. SM volume and age were positive predictors of pCR disease in both male and female patients following NAT for LARC. Future studies investigating the impact of body composition on clinical outcomes and patients on other neoadjuvant regimens such as TNT are potential avenues for further research.

The association of body composition on chemotherapy toxicities in non-metastatic colorectal cancer patients: a systematic review.

BACKGROUND: In recent years, certain body composition measures, assessed by computed tomography (CT), have been found to be associated with chemotherapy toxicities. This review aims to explore available data on the relationship between skeletal muscle and adiposity, including visceral adipose tissue (VAT), subcutaneous adipose tissue (SAT), intramuscular and intermuscular adipose tissue and their association with chemotherapy toxicity in non-metastatic colorectal cancer (CRC) patients. METHODS: A systematic literature search following PRISMA guidelines was conducted in Medline, Embase, Cochrane and Web of Science, for papers published between 2011 and 2023. The search strategy combined keywords and MESH terms relevant to 'body composition', 'chemotherapy toxicities', and 'non-metastatic colorectal cancer'. RESULTS: Out of 3868 studies identified, six retrospective studies fulfilled the inclusion criteria with 1024 eligible patients. Low skeletal muscle mass was strongly associated with increased incidence of both chemotherapy toxicities and dose-limiting toxicity (DLT). The association of VAT, intramuscular and intermuscular adiposity was heterogeneous and inconclusive. There was no association between SAT and chemotherapy intolerance. No universal definitions or cut-offs for sarcopenia and obesity were noted. All studies utilized 2-dimensional (2D) CT slices for CT body composition assessment with varied selection on the vertebral landmark and inconsistent reporting of tissue-defining Hounsfield unit (HU) measurements. CONCLUSION: Low skeletal muscle is associated with chemotherapy toxicities in non-metastatic CRC. However, quality evidence on the role of adiposity is limited and heterogeneous. More studies are needed to confirm these associations with an emphasis on a more coherent body composition definition and an approach to its assessment, especially regarding sarcopenia.

Identification of Differences in Body Composition Measures Using 3D-Derived Artificial Intelligence from Multiple CT Scans across the L3 Vertebra Compared to a Single Mid-Point L3 CT Scan.

Purpose: Body composition analysis in colorectal cancer (CRC) typically utilises a single 2D-abdominal axial CT slice taken at the mid-L3 level. The use of artificial intelligence (AI) allows for analysis of the entire L3 vertebra (non-mid-L3 and mid-L3). The goal of this study was to determine if the use of an AI approach offered any additional information on capturing body composition measures. Methods: A total of 2203 axial CT slices of the entire L3 level (4-46 slices were available per patient) were retrospectively collected from 203 CRC patients treated at Western Health, Melbourne (97 males; 47.8%). A pretrained artificial intelligence (AI) model was used to segment muscle, visceral adipose tissue (VAT), and subcutaneous adipose tissue (SAT) on these slices. The difference in body composition measures between mid-L3 and non-mid-L3 scans was compared for each patient, and for males and females separately. Results: Body composition measures derived from non-mid-L3 scans exhibited a median range of 0.85% to 6.28% (average percent difference) when compared to the use of a single mid-L3 scan. Significant variation in the VAT surface area (p = 0.02) was observed in females compared to males, whereas male patients exhibited a greater variation in SAT surface area (p < 0.001) and radiodensity (p = 0.007). Conclusion: Significant differences in various body composition measures were observed when comparing non-mid-L3 slices to only the mid-L3 slice. Researchers should be aware that considering only the use of a single midpoint L3 CT scan slice will impact the estimate of body composition measurements.

Vision-Related Quality of Life and Ocular Parameters in End-Stage Renal Disease Patients Undergoing Hemodialysis.

Purpose: To evaluate the vision-related quality of life (VRQoL) of patients receiving hemodialysis through the assessment of the impact of vision impairment (IVI) questionnaire and ocular parameters, including best-corrected visual acuity (BCVA), intraocular pressure (IOP), and refraction as calculated by spherical equivalent (SE) of each eye. Methods: Fifty-one patients with end-stage renal disease undergoing hemodialysis at a single center were recruited, and a total of 77 eyes were evaluated. BCVA, IOP, and SE were evaluated before and after hemodialysis (within 30 min). Results: Of the 51 patients recruited, 13 (25%) were female, 37 (73%) were male, and one (2%) chose not to specify gender. The mean age was 61.85 ± 32 years. The mobility IVI score was correlated significantly with the presence of hypertension (P = 0.01), eye drop usage (P = 0.04), and gender (P = 0.04). Emotional IVI scores were correlated significantly with diabetes (P = 0.03) and hypertension (P < 0.01). IOP significantly correlated with the IVI overall score (P = 0.02), including the reading IVI subscale and the emotional IVI subscale. Several factors were associated with posthemodialysis ocular parameters, including predialysis ocular parameters, age, and hypertension (P < 0.05 for all). Conclusions: IOP significantly correlated with VRQoL in hemodialysis patients. Demographic variables such as diabetes status, hypertension, eye drop usage, and gender also significantly correlated with subsections of the IVI questionnaire. This study investigated the relationship between ocular parameters and VRQoL in hemodialysis patients, and future longitudinal research is needed to further elucidate the mechanisms.

Can AI-based body composition assessment outperform body surface area in predicting dose-limiting toxicities for colonic cancer patients on chemotherapy?

PURPOSE: Gold standard chemotherapy dosage is based on body surface area (BSA); however many patients experience dose-limiting toxicities (DLT). We aimed to evaluate the effectiveness of BSA, two-dimensional (2D) and three-dimensional (3D) body composition (BC) measurements derived from Lumbar 3 vertebra (L3) computed tomography (CT) slices, in predicting DLT in colon cancer patients. METHODS: 203 patients (60.87 ± 12.42 years; 97 males, 47.8%) receiving adjuvant chemotherapy (Oxaliplatin and/or 5-Fluorouracil) were retrospectively evaluated. An artificial intelligence segmentation model was used to extract 2D and 3D body composition measurements from each patients' single mid-L3 CT slice as well as multiple-L3 CT scans to produce a 3D BC report. DLT was defined as any incidence of dose reduction or discontinuation due to chemotherapy toxicities. A receiver operating characteristic (ROC) analysis was performed on BSA and individual body composition measurements to demonstrate their predictive performance. RESULTS: A total of 120 (59.1%) patients experienced DLT. Age and BSA did not vary significantly between DLT and non-DLT group. Females were significantly more likely to experience DLT (p = 4.9 × 10-3). In all patients, the predictive effectiveness of 2D body composition measurements (females: AUC = 0.50-0.54; males: AUC = 0.50-0.61) was equivalent to that of BSA (females: AUC = 0.49; males: AUC = 0.58). The L3 3D skeletal muscle volume was the most predictive indicator of DLT (AUC of 0.66 in females and 0.64 in males). CONCLUSION: Compared to BSA and 2D body composition measurements, 3D L3 body composition measurements had greater potential to predict DLT in CRC patients receiving chemotherapy and this was sex dependent.

Keratoconus International Consortium (KIC)- advancing keratoconus research.

CLINICAL RELEVANCE: The Keratoconus International Consortium (KIC) will allow better understanding of keratoconus. BACKGROUND: Keratoconus is a disorder characterised by corneal elevation and thinning, leading to reduced vision. The current gaps in understanding of this disease will be discussed and the need for a multi-pronged and multi-centre engagement to enhance our understanding of keratoconus will be highlighted. DESIGN: KIC has been established to address the gaps in our understanding of keratoconus with the aim of collecting baseline as well as longitudinal data on several fields. PARTICIPANTS: Keratoconus and control (no corneal condition) subjects from different sites globally will be recruited in the study. METHODS: KIC collects data using an online, secure database, which enables standardised data collection at member sites. Data fields collected include medical history, clinical features, quality of life and economic burden questionnaires and possible genetic sample collection from patients of different ethnicities across different geographical locations. RESULTS: There are currently 40 Australian and international clinics or hospital departments who have joined the KIC. Baseline data has so far been collected on 1130 keratoconus patients and indicates a median age of 29.70 years with 61% being male. A total of 15.3% report a positive family history of keratoconus and 57.7% self-report a history of frequent eye rubbing. CONCLUSION: The strength of this consortium is its international, collaborative design and use of a common data collection tool. Inclusion and analyses of cross-sectional and longitudinal data will help answer many questions that remain in keratoconus, including factors affecting progression and treatment outcomes.

Can sarcopenia predict survival in locally advanced rectal cancer patients?

BACKGROUND: There is mounting evidence that suggests sarcopenia can be used to predict survival outcomes in patients with colon cancer. However, the effect on locally advanced rectal cancer (LARC) is less clear. We sought to determine the association between sarcopenia on Overall Survival and Recurrence-free Survival (OS and RFS) in patients with LARC undergoing multimodal treatment. METHODS: A retrospective study was undertaken of all pre-treatment stage 2-3 rectal cancer patients who underwent neo-adjuvant treatment and surgery with curative intent between January 2010 and September 2016 at Western Health. Sarcopenia was measured on pre-treatment staging scans at the third lumbar vertebrae and defined using cohort-derived, sex-specific thresholds. Primary outcomes were OS and RFS. RESULTS: A total of 132 patients with LARC were analysed. Sarcopenia: Hazard ratio (HR) 3.71; 95% CI, 1.28-10.75, P = 0.016 was independently associated with worse Overall Survival following multivariate analysis. There was no significant relationship between sarcopenia and RFS: Time ratio (TR) 1.67; 95% CI 0.52-5.34, P = 0.386. CONCLUSION: Sarcopenia was found to be an independent risk factor for worse overall survival, but not recurrence free survival, in patients with locally advanced rectal cancer undergoing neo-adjuvant chemo-radiotherapy and surgery with curative intent.

The extracellular microenvironment in immune dysregulation and inflammation in retinal disorders.

Inherited retinal dystrophies (IRDs) as well as genetically complex retinal phenotypes represent a heterogenous group of ocular diseases, both on account of their phenotypic and genotypic characteristics. Therefore, overlaps in clinical features often complicate or even impede their correct clinical diagnosis. Deciphering the molecular basis of retinal diseases has not only aided in their disease classification but also helped in our understanding of how different molecular pathologies may share common pathomechanisms. In particular, these relate to dysregulation of two key processes that contribute to cellular integrity, namely extracellular matrix (ECM) homeostasis and inflammation. Pathological changes in the ECM of Bruch's membrane have been described in both monogenic IRDs, such as Sorsby fundus dystrophy (SFD) and Doyne honeycomb retinal dystrophy (DHRD), as well as in the genetically complex age-related macular degeneration (AMD) or diabetic retinopathy (DR). Additionally, complement system dysfunction and distorted immune regulation may also represent a common connection between some IRDs and complex retinal degenerations. Through highlighting such overlaps in molecular pathology, this review aims to illuminate how inflammatory processes and ECM homeostasis are linked in the healthy retina and how their interplay may be disturbed in aging as well as in disease.

Comparison of CT derived body composition at the thoracic T4 and T12 with lumbar L3 vertebral levels and their utility in patients with rectal cancer.

BACKGROUND: Computed tomography (CT) derived body composition measurements of sarcopenia are an emerging form of prognostication in many disease processes. Although the L3 vertebral level is commonly used to measure skeletal muscle mass, other studies have suggested the utilisation of other segments. This study was performed to assess the variation and reproducibility of skeletal muscle mass at vertebral levels T4, T12 and L3 in pre-operative rectal cancer patients. If thoracic measurements were equivalent to those at L3, it will allow for body composition comparisons in a larger range of cancers where lumbar CT images are not routinely measured. RESEARCH METHODS: Patients with stage I - III rectal cancer, undergoing curative resection from 2010 - 2014, were assessed. CT based quantification of skeletal muscle was used to determine skeletal muscle cross sectional area (CSA) and skeletal muscle index (SMI). Systematic differences between the measurements at L3 with T4 and T12 vertebral levels were evaluated by percentile rank differences to assess distribution of differences and ordinary least product regression (OLP) to detect and distinguish fixed and proportional bias. RESULTS: Eighty eligible adult patients were included. Distribution of differences between T12 SMI and L3 SMI were more marked than differences between T4 SMI and L3 SMI. There was no fix or proportional bias with T4 SMI, but proportional bias was detected with T12 SMI measurements. T4 CSA duplicate measurements had higher test-retest reliability: coefficient of repeatability was 34.10 cm2 for T4 CSA vs 76.00 cm2 for T12 CSA. Annotation time (minutes) with L3 as reference, the median difference was 0.85 for T4 measurements and -0.03 for T12 measurements. Thirty-seven patients (46%) had evidence of sarcopenia at the L3 vertebral level, with males exhibiting higher rates of sarcopenia. However, there was no association between sarcopenia and post-operative complications, recurrence or hospital LOS (length of stay) in patients undergoing curative resection. CONCLUSIONS: Quantifying skeletal muscle mass at the T4 vertebral level is comparable to measures achieved at L3 in patients with rectal cancer, notwithstanding annotation time for T4 measurements are longer.

Non-genetic risk factors for keratoconus.

Keratoconus is a complex and multifactorial disease and its exact aetiology remains unknown. This current study examined the important environmental risk factors and their association with keratoconus. This study was registered in the PROSPERO International Prospective Register of systematic reviews under registration number CRD42021256792 in 2021. Scopus, Web of Science, PubMed, and Cochrane CENTRAL databases were searched for all relevant articles published from 1 January 1900 to 31 July 2021. National Institutes of Health Quality Assessment Tool was used to assess the methodological quality of the studies. The assessment for statistical heterogeneity was assessed using the Z-statistics on RevMan v5.4. P-value of <0.05 was considered as statistically significant and I2 < 25% as homogenous. Thirty studies were included in this meta-analysis. Pooled odds ratio was calculated with 95% CI. The pooled odds ratio (OR) of eye rubbing, atopy, asthma, and eczema was 3.64 (95% CI, 2.02, 6.57), 1.90 (95% CI, 1.22, 2.94), 1.36 (95% CI, 1.15, 1.61) and 1.90 (95% CI, 1.22, 2.94), respectively. The OR for diabetes was 0.86 (95% CI 0.73, 1.02), and use of sunglasses, contact lens, allergic conjunctivitis, side sleep position and prone sleep position was 0.40 (95% CI, 0.16, 0.99), 1.68 (0.70, 4.00), 2.24 (95% CI, 0.68, 7.36), 3.81 (95% CI, 0.31, 46.23), 12.76 (95% CI, 0.27, 598.58), respectively. Twenty studies were considered to be of high quality, nine to be moderate and one to be low. Environmental risk factors have been identified to play a role in the susceptibility of keratoconus. However, further large-scale longitudinal studies are needed to understand the mechanisms between environmental risk factors and keratoconus.

Exploring the contribution of ARMS2 and HTRA1 genetic risk factors in age-related macular degeneration.

Age-related macular degeneration (AMD) is the leading cause of severe irreversible central vision loss in individuals over 65 years old. Genome-wide association studies (GWASs) have shown that the region at chromosome 10q26, where the age-related maculopathy susceptibility (ARMS2/LOC387715) and HtrA serine peptidase 1 (HTRA1) genes are located, represents one of the strongest associated loci for AMD. However, the underlying biological mechanism of this genetic association has remained elusive. In this article, we extensively review the literature by us and others regarding the ARMS2/HTRA1 risk alleles and their functional significance. We also review the literature regarding the presumed function of the ARMS2 protein and the molecular processes of the HTRA1 protein in AMD pathogenesis in vitro and in vivo, including those of transgenic mice overexpressing HtrA1/HTRA1 which developed Bruch's membrane (BM) damage, choroidal neovascularization (CNV), and polypoidal choroidal vasculopathy (PCV), similar to human AMD patients. The elucidation of the molecular mechanisms of the ARMS2 and HTRA1 susceptibility loci has begun to untangle the complex biological pathways underlying AMD pathophysiology, pointing to new testable paradigms for treatment.

Accuracy of Machine Learning Assisted Detection of Keratoconus: A Systematic Review and Meta-Analysis.

(1) Background: The objective of this review was to synthesize available data on the use of machine learning to evaluate its accuracy (as determined by pooled sensitivity and specificity) in detecting keratoconus (KC), and measure reporting completeness of machine learning models in KC based on TRIPOD (the transparent reporting of multivariable prediction models for individual prognosis or diagnosis) statement. (2) Methods: Two independent reviewers searched the electronic databases for all potential articles on machine learning and KC published prior to 2021. The TRIPOD 29-item checklist was used to evaluate the adherence to reporting guidelines of the studies, and the adherence rate to each item was computed. We conducted a meta-analysis to determine the pooled sensitivity and specificity of machine learning models for detecting KC. (3) Results: Thirty-five studies were included in this review. Thirty studies evaluated machine learning models for detecting KC eyes from controls and 14 studies evaluated machine learning models for detecting early KC eyes from controls. The pooled sensitivity for detecting KC was 0.970 (95% CI 0.949-0.982), with a pooled specificity of 0.985 (95% CI 0.971-0.993), whereas the pooled sensitivity of detecting early KC was 0.882 (95% CI 0.822-0.923), with a pooled specificity of 0.947 (95% CI 0.914-0.967). Between 3% and 48% of TRIPOD items were adhered to in studies, and the average (median) adherence rate for a single TRIPOD item was 23% across all studies. (4) Conclusions: Application of machine learning model has the potential to make the diagnosis and monitoring of KC more efficient, resulting in reduced vision loss to the patients. This review provides current information on the machine learning models that have been developed for detecting KC and early KC. Presently, the machine learning models performed poorly in identifying early KC from control eyes and many of these research studies did not follow established reporting standards, thus resulting in the failure of these clinical translation of these machine learning models. We present possible approaches for future studies for improvement in studies related to both KC and early KC models to more efficiently and widely utilize machine learning models for diagnostic process.

Examining Corneal Tissue Exportation Fee and Its Impact on Equitable Allocation.

METHODS: We conducted grounded theory semistructured interviews, purposively inviting participants until themed saturation was met. Sentiment analysis was used to determine opinion. RESULTS: We interviewed n = 92 global eye tissue and eye bank professionals. We determined that corneal tissue, which is exported, costs between US $100 and US $6000 or is provided as gratis. Collectively, interviewees indicated that, globally, there were no fixed fee structures in place, and the fee was influenced by multiple factors on both export and import sides. They indicated that ultimately corneas were allocated based on the importers' ability to pay the price determined by the exporting eye bank. DISCUSSION: Allocation of corneal tissue, which is exported, is influenced by the fees charged by the exporters to meet their bottom line and the funds available to importers. Therefore, export allocation is not equitable, with those who can pay a higher fee, prioritized. Steps to guide and support exporters with the development of fee structures that promote equitable allocation are essential. This will assist both export and import eye bank development, corneal tissue access development, and those awaiting a corneal transplant.

Identification of novel loci influencing refractive error in East Asian populations using an extreme phenotype design.

The global "myopia boom" has raised significant international concerns. Despite a higher myopia prevalence in Asia, previous large-scale genome-wide association studies (GWASs) were mostly based on European descendants. Here, we report a GWAS of spherical equivalent (SE) in 1852 Chinese Han individuals with extreme SE from Guangzhou (631 < -6.00D and 574 > 0.00D) and Wenzhou (593 < -6.00D and 54 > -1.75D), followed by a replication study in two independent cohorts with totaling 3538 East Asian individuals. The discovery GWAS and meta-analysis identify three novel loci, which show genome-wide significant associations with SE, including 1q25.2 FAM163A, 10p11.22 NRP1/PRAD3, and 10p11.21 ANKRD30A/MTRNR2L7, together explaining 3.34% of SE variance. 10p11.21 is successfully replicated. The allele frequencies of all three loci show significant differences between major continental groups (P < 0.001). The SE reducing (more myopic) allele of rs10913877 (1q25.2 FAM163A) demonstrates the highest frequency in East Asians and much lower frequencies in Europeans and Africans (EAS = 0.60, EUR = 0.20, and AFR = 0.18). The gene-based analysis additionally identifies three novel genes associated with SE, including EI24, LHX5, and ARPP19. These results provide new insights into myopia pathogenesis and indicate the role of genetic heterogeneity in myopia epidemiology among different ethnicities.

Machine learning with a reduced dimensionality representation of comprehensive Pentacam tomography parameters to identify subclinical keratoconus.

PURPOSE: To investigate the performance of a machine learning model based on a reduced dimensionality parameter space derived from complete Pentacam parameters to identify subclinical keratoconus (KC). METHODS: All 1692 available parameters were obtained from the Pentacam imaging machine on 145 subclinical KC and 122 control eyes. We applied a principal component analysis (PCA) to the complete Pentacam dataset to reduce its parameter dimensionality. Subsequently, we investigated machine learning performance of the random forest algorithm with increasing numbers of components to identify their optimal number for detecting subclinical KC from control eyes. RESULTS: The dimensionality of the complete set of 1692 Pentacam parameters was reduced to 267 principal components using PCA. Subsequent selection of 15 of these principal components explained over 85% of the variance of the original Pentacam-derived parameters and input to train a random forest machine learning model to achieve the best accuracy of 98% in detecting subclinical KC eyes. The model established also reached a high sensitivity of 97% in identification of subclinical KC and a specificity of 98% in recognizing control eyes. CONCLUSIONS: A random forest-based model trained using a modest number of components derived from a reduced dimensionality representation of complete Pentacam system parameters allowed for high accuracy of subclinical KC identification.

Deep Learning Applied to Automated Segmentation of Geographic Atrophy in Fundus Autofluorescence Images.

Purpose: This study describes the development of a deep learning algorithm based on the U-Net architecture for automated segmentation of geographic atrophy (GA) lesions in fundus autofluorescence (FAF) images. Methods: Image preprocessing and normalization by modified adaptive histogram equalization were used for image standardization to improve effectiveness of deep learning. A U-Net-based deep learning algorithm was developed and trained and tested by fivefold cross-validation using FAF images from clinical datasets. The following metrics were used for evaluating the performance for lesion segmentation in GA: dice similarity coefficient (DSC), DSC loss, sensitivity, specificity, mean absolute error (MAE), accuracy, recall, and precision. Results: In total, 702 FAF images from 51 patients were analyzed. After fivefold cross-validation for lesion segmentation, the average training and validation scores were found for the most important metric, DSC (0.9874 and 0.9779), for accuracy (0.9912 and 0.9815), for sensitivity (0.9955 and 0.9928), and for specificity (0.8686 and 0.7261). Scores for testing were all similar to the validation scores. The algorithm segmented GA lesions six times more quickly than human performance. Conclusions: The deep learning algorithm can be implemented using clinical data with a very high level of performance for lesion segmentation. Automation of diagnostics for GA assessment has the potential to provide savings with respect to patient visit duration, operational cost and measurement reliability in routine GA assessments. Translational Relevance: A deep learning algorithm based on the U-Net architecture and image preprocessing appears to be suitable for automated segmentation of GA lesions on clinical data, producing fast and accurate results.

Corneal supply and the use of technology to reduce its demand: A review.

Recovery and access to end-of-life corneal tissue for corneal transplantation, training and research is globally maldistributed. The reasons for the maldistribution are complex and multifaceted, and not well defined or understood. Currently there are few solutions available to effectively address these issues. This review provides an overview of the system, key issues impacting recovery and allocation and emphasises how end-user ophthalmologists and researchers, with support from administrators and the wider sector, can assist in increasing access long-term through sustaining eye banks nationally and globally. We posit that prevention measures and improved surgical techniques, together with the development of novel therapies will play a significant role in reducing demand and enhance the equitable allocation of corneas.

Model Structure Uncertainty in the Characterization and Growth of Geographic Atrophy.

Purpose: To identify the most suitable model for assessing the rate of growth of total geographic atrophy (GA) by analysis of model structure uncertainty. Methods: Model structure uncertainty refers to unexplained variability arising from the choice of mathematical model and represents an example of epistemic uncertainty. In this study, we quantified this uncertainty to help identify a model most representative of GA progression. Fundus autofluorescence (FAF) images and GA progression data (i.e., total GA area estimation at each presentation) were acquired using Spectralis HRA+OCT instrumentation and RegionFinder software. Six regression models were evaluated. Models were compared using various statistical tests, [i.e., coefficient of determination (r2), uncertainty metric (U), and test of significance for the correlation coefficient, r], as well as adherence to expected physical and clinical assumptions of GA growth. Results: Analysis was carried out for 81 GA-affected eyes, 531 FAF images (range: 3-17 images per eye), over median of 57 months (IQR: 42, 74), with a mean baseline lesion size of 2.62 ± 4.49 mm2 (range: 0.11-20.69 mm2). The linear model proved to be the most representative of total GA growth, with lowest average uncertainty (original scale: U = 0.025, square root scale: U = 0.014), high average r2 (original scale: 0.92, square root scale: 0.93), and applicability of the model was supported by a high correlation coefficient, r, with statistical significance (P = 0.01). Conclusions: Statistical analysis of uncertainty suggests that the linear model provides an effective and practical representation of the rate and progression of total GA growth based on data from patient presentations in clinical settings. Translational Relevance: Identification of correct model structure to characterize rate of growth of total GA in the retina using FAF images provides an objective metric for comparing interventions and charting GA progression in clinical presentations.