AIROGS: Artificial Intelligence for Robust Glaucoma Screening Challenge.

The early detection of glaucoma is essential in preventing visual impairment. Artificial intelligence (AI) can be used to analyze color fundus photographs (CFPs) in a cost-effective manner, making glaucoma screening more accessible. While AI models for glaucoma screening from CFPs have shown promising results in laboratory settings, their performance decreases significantly in real-world scenarios due to the presence of out-of-distribution and low-quality images. To address this issue, we propose the Artificial Intelligence for Robust Glaucoma Screening (AIROGS) challenge. This challenge includes a large dataset of around 113,000 images from about 60,000 patients and 500 different screening centers, and encourages the development of algorithms that are robust to ungradable and unexpected input data. We evaluated solutions from 14 teams in this paper and found that the best teams performed similarly to a set of 20 expert ophthalmologists and optometrists. The highest-scoring team achieved an area under the receiver operating characteristic curve of 0.99 (95% CI: 0.98-0.99) for detecting ungradable images on-the-fly. Additionally, many of the algorithms showed robust performance when tested on three other publicly available datasets. These results demonstrate the feasibility of robust AI-enabled glaucoma screening.

CHILDSTAR: CHIldren Living With Diabetes See and Thrive with AI Review.

Background: Artificial intelligence (AI) appears capable of detecting diabetic retinopathy (DR) with a high degree of accuracy in adults; however, there are few studies in children and young adults. Methods: Children and young adults (3-26 years) with type 1 diabetes mellitus (T1DM) or type 2 diabetes mellitus (T2DM) were screened at the Dhaka BIRDEM-2 hospital, Bangladesh. All gradable fundus images were uploaded to Cybersight AI for interpretation. Two main outcomes were considered at a patient level: 1) Any DR, defined as mild non-proliferative diabetic retinopathy (NPDR or more severe; and 2) Referable DR, defined as moderate NPDR or more severe. Diagnostic test performance comparing Orbis International's Cybersight AI with the reference standard, a fully qualified optometrist certified in DR grading, was assessed using the Matthews correlation coefficient (MCC), area under the receiver operating characteristic curve (AUC-ROC), area under the precision-recall curve (AUC-PR), sensitivity, specificity, positive and negative predictive values. Results: Among 1274 participants (53.1% female, mean age 16.7 years), 19.4% (n = 247) had any DR according to AI. For referable DR, 2.35% (n = 30) were detected by AI. The sensitivity and specificity of AI for any DR were 75.5% (CI 69.7-81.3%) and 91.8% (CI 90.2-93.5%) respectively, and for referable DR, these values were 84.2% (CI 67.8-100%) and 98.9% (CI 98.3%-99.5%). The MCC, AUC-ROC and the AUC-PR for referable DR were 63.4, 91.2 and 76.2% respectively. AI was most successful in accurately classifying younger children with shorter duration of diabetes. Conclusions: Cybersight AI accurately detected any DR and referable DR among children and young adults, despite its algorithms having been trained on adults. The observed high specificity is particularly important to avoid over-referral in low-resource settings. AI may be an effective tool to reduce demands on scarce physician resources for the care of children with diabetes in low-resource settings.

Feasibility and acceptance of artificial intelligence-based diabetic retinopathy screening in Rwanda.

BACKGROUND: Evidence on the practical application of artificial intelligence (AI)-based diabetic retinopathy (DR) screening is needed. METHODS: Consented participants were screened for DR using retinal imaging with AI interpretation from March 2021 to June 2021 at four diabetes clinics in Rwanda. Additionally, images were graded by a UK National Health System-certified retinal image grader. DR grades based on the International Classification of Diabetic Retinopathy with a grade of 2.0 or higher were considered referable. The AI system was designed to detect optic nerve and macular anomalies outside of DR. A vertical cup to disc ratio of 0.7 and higher and/or macular anomalies recognised at a cut-off of 60% and higher were also considered referable by AI. RESULTS: Among 827 participants (59.6% women (n=493)) screened by AI, 33.2% (n=275) were referred for follow-up. Satisfaction with AI screening was high (99.5%, n=823), and 63.7% of participants (n=527) preferred AI over human grading. Compared with human grading, the sensitivity of the AI for referable DR was 92% (95% CI 0.863%, 0.968%), with a specificity of 85% (95% CI 0.751%, 0.882%). Of the participants referred by AI: 88 (32.0%) were for DR only, 109 (39.6%) for DR and an anomaly, 65 (23.6%) for an anomaly only and 13 (4.73%) for other reasons. Adherence to referrals was highest for those referred for DR at 53.4%. CONCLUSION: DR screening using AI led to accurate referrals from diabetes clinics in Rwanda and high rates of participant satisfaction, suggesting AI screening for DR is practical and acceptable.

Impact of Artificial Intelligence Assessment of Diabetic Retinopathy on Referral Service Uptake in a Low-Resource Setting: The RAIDERS Randomized Trial.

Purpose: This trial was designed to determine if artificial intelligence (AI)-supported diabetic retinopathy (DR) screening improved referral uptake in Rwanda. Design: The Rwanda Artificial Intelligence for Diabetic Retinopathy Screening (RAIDERS) study was an investigator-masked, parallel-group randomized controlled trial. Participants: Patients ≥ 18 years of age with known diabetes who required referral for DR based on AI interpretation. Methods: The RAIDERS study screened for DR using retinal imaging with AI interpretation implemented at 4 facilities from March 2021 through July 2021. Eligible participants were assigned randomly (1:1) to immediate feedback of AI grading (intervention) or communication of referral advice after human grading was completed 3 to 5 days after the initial screening (control). Main Outcome Measures: Difference between study groups in the rate of presentation for referral services within 30 days of being informed of the need for a referral visit. Results: Of the 823 clinic patients who met inclusion criteria, 275 participants (33.4%) showed positive findings for referable DR based on AI screening and were randomized for inclusion in the trial. Study participants (mean age, 50.7 years; 58.2% women) were randomized to the intervention (n = 136 [49.5%]) or control (n = 139 [50.5%]) groups. No significant intergroup differences were found at baseline, and main outcome data were available for analyses for 100% of participants. Referral adherence was statistically significantly higher in the intervention group (70/136 [51.5%]) versus the control group (55/139 [39.6%]; P = 0.048), a 30.1% increase. Older age (odds ratio [OR], 1.04; 95% confidence interval [CI], 1.02-1.05; P < 0.0001), male sex (OR, 2.07; 95% CI, 1.22-3.51; P = 0.007), rural residence (OR, 1.79; 95% CI, 1.07-3.01; P = 0.027), and intervention group (OR, 1.74; 95% CI, 1.05-2.88; P = 0.031) were statistically significantly associated with acceptance of referral in multivariate analyses. Conclusions: Immediate feedback on referral status based on AI-supported screening was associated with statistically significantly higher referral adherence compared with delayed communications of results from human graders. These results provide evidence for an important benefit of AI screening in promoting adherence to prescribed treatment for diabetic eye care in sub-Saharan Africa.

The Role of Artificial Intelligence in the Diagnosis and Management of Glaucoma.

Glaucomatous optic neuropathy is the leading cause of irreversible blindness worldwide. Diagnosis and monitoring of disease involves integrating information from the clinical examination with subjective data from visual field testing and objective biometric data that includes pachymetry, corneal hysteresis, and optic nerve and retinal imaging. This intricate process is further complicated by the lack of clear definitions for the presence and progression of glaucomatous optic neuropathy, which makes it vulnerable to clinician interpretation error. Artificial intelligence (AI) and AI-enabled workflows have been proposed as a plausible solution. Applications derived from this field of computer science can improve the quality and robustness of insights obtained from clinical data that can enhance the clinician's approach to patient care. This review clarifies key terms and concepts used in AI literature, discusses the current advances of AI in glaucoma, elucidates the clinical advantages and challenges to implementing this technology, and highlights potential future applications.

DISEASE CLASSIFICATION OF MACULAR OPTICAL COHERENCE TOMOGRAPHY SCANS USING DEEP LEARNING SOFTWARE: Validation on Independent, Multicenter Data.

PURPOSE: To evaluate Pegasus optical coherence tomography (OCT), a clinical decision support software for the identification of features of retinal disease from macula OCT scans, across heterogenous populations involving varying patient demographics, device manufacturers, acquisition sites, and operators. METHODS: Five thousand five hundred and eighty-eight normal and anomalous macular OCT volumes (162,721 B-scans), acquired at independent centers in five countries, were processed using the software. Results were evaluated against ground truth provided by the data set owners. RESULTS: Pegasus-OCT performed with areas under the curve of the receiver operating characteristic of at least 98% for all data sets in the detection of general macular anomalies. For scans of sufficient quality, the areas under the curve of the receiver operating characteristic for general age-related macular degeneration and diabetic macular edema detection were found to be at least 99% and 98%, respectively. CONCLUSION: The ability of a clinical decision support system to cater for different populations is key to its adoption. Pegasus-OCT was shown to be able to detect age-related macular degeneration, diabetic macular edema, and general anomalies in OCT volumes acquired across multiple independent sites with high performance. Its use thus offers substantial promise, with the potential to alleviate the burden of growing demand in eye care services caused by retinal disease.

Evaluation of an AI system for the automated detection of glaucoma from stereoscopic optic disc photographs: the European Optic Disc Assessment Study.

OBJECTIVES: To evaluate the performance of a deep learning based Artificial Intelligence (AI) software for detection of glaucoma from stereoscopic optic disc photographs, and to compare this performance to the performance of a large cohort of ophthalmologists and optometrists. METHODS: A retrospective study evaluating the diagnostic performance of an AI software (Pegasus v1.0, Visulytix Ltd., London UK) and comparing it with that of 243 European ophthalmologists and 208 British optometrists, as determined in previous studies, for the detection of glaucomatous optic neuropathy from 94 scanned stereoscopic photographic slides scanned into digital format. RESULTS: Pegasus was able to detect glaucomatous optic neuropathy with an accuracy of 83.4% (95% CI: 77.5-89.2). This is comparable to an average ophthalmologist accuracy of 80.5% (95% CI: 67.2-93.8) and average optometrist accuracy of 80% (95% CI: 67-88) on the same images. In addition, the AI system had an intra-observer agreement (Cohen's Kappa, κ) of 0.74 (95% CI: 0.63-0.85), compared with 0.70 (range: -0.13-1.00; 95% CI: 0.67-0.73) and 0.71 (range: 0.08-1.00) for ophthalmologists and optometrists, respectively. There was no statistically significant difference between the performance of the deep learning system and ophthalmologists or optometrists. CONCLUSION: The AI system obtained a diagnostic performance and repeatability comparable to that of the ophthalmologists and optometrists. We conclude that deep learning based AI systems, such as Pegasus, demonstrate significant promise in the assisted detection of glaucomatous optic neuropathy.

Detection of concealed cars in complex cargo X-ray imagery using Deep Learning.

BACKGROUND: Non-intrusive inspection systems based on X-ray radiography techniques are routinely used at transport hubs to ensure the conformity of cargo content with the supplied shipping manifest. As trade volumes increase and regulations become more stringent, manual inspection by trained operators is less and less viable due to low throughput. Machine vision techniques can assist operators in their task by automating parts of the inspection workflow. Since cars are routinely involved in trafficking, export fraud, and tax evasion schemes, they represent an attractive target for automated detection and flagging for subsequent inspection by operators. OBJECTIVE: Development and evaluation of a novel method for the automated detection of cars in complex X-ray cargo imagery. METHODS: X-ray cargo images from a stream-of-commerce dataset were classified using a window-based scheme. The limited number of car images was addressed by using an oversampling scheme. Different Convolutional Neural Network (CNN) architectures were compared with well-established bag of words approaches. In addition, robustness to concealment was evaluated by projection of objects into car images. RESULTS: CNN approaches outperformed all other methods evaluated, achieving 100% car image classification rate for a false positive rate of 1-in-454. Cars that were partially or completely obscured by other goods, a modus operandi frequently adopted by criminals, were correctly detected. CONCLUSIONS: We believe that this level of performance suggests that the method is suitable for deployment in the field. It is expected that the generic object detection workflow described can be extended to other object classes given the availability of suitable training data.

Automated X-ray image analysis for cargo security: Critical review and future promise.

We review the relatively immature field of automated image analysis for X-ray cargo imagery. There is increasing demand for automated analysis methods that can assist in the inspection and selection of containers, due to the ever-growing volumes of traded cargo and the increasing concerns that customs- and security-related threats are being smuggled across borders by organised crime and terrorist networks. We split the field into the classical pipeline of image preprocessing and image understanding. Preprocessing includes: image manipulation; quality improvement; Threat Image Projection (TIP); and material discrimination and segmentation. Image understanding includes: Automated Threat Detection (ATD); and Automated Contents Verification (ACV). We identify several gaps in the literature that need to be addressed and propose ideas for future research. Where the current literature is sparse we borrow from the single-view, multi-view, and CT X-ray baggage domains, which have some characteristics in common with X-ray cargo.

Real-time monitoring of specific oxygen uptake rates of embryonic stem cells in a microfluidic cell culture device.

Oxygen plays a key role in stem cell biology as a signaling molecule and as an indicator of cell energy metabolism. Quantification of cellular oxygen kinetics, i.e. the determination of specific oxygen uptake rates (sOURs), is routinely used to understand metabolic shifts. However current methods to determine sOUR in adherent cell cultures rely on cell sampling, which impacts on cellular phenotype. We present real-time monitoring of cell growth from phase contrast microscopy images, and of respiration using optical sensors for dissolved oxygen. Time-course data for bulk and peri-cellular oxygen concentrations obtained for Chinese hamster ovary (CHO) and mouse embryonic stem cell (mESCs) cultures successfully demonstrated this non-invasive and label-free approach. Additionally, we confirmed non-invasive detection of cellular responses to rapidly changing culture conditions by exposing the cells to mitochondrial inhibiting and uncoupling agents. For the CHO and mESCs, sOUR values between 8 and 60 amol cell(-1) s(-1) , and 5 and 35 amol cell(-1) s(-1) were obtained, respectively. These values compare favorably with literature data. The capability to monitor oxygen tensions, cell growth, and sOUR, of adherent stem cell cultures, non-invasively and in real time, will be of significant benefit for future studies in stem cell biology and stem cell-based therapies.

Automated and online characterization of adherent cell culture growth in a microfabricated bioreactor.

Adherent cell lines are widely used across all fields of biology, including drug discovery, toxicity studies, and regenerative medicine. However, adherent cell processes are often limited by a lack of advances in cell culture systems. While suspension culture processes benefit from decades of development of instrumented bioreactors, adherent cultures are typically performed in static, noninstrumented flasks and well-plates. We previously described a microfabricated bioreactor that enables a high degree of control on the microenvironment of the cells while remaining compatible with standard cell culture protocols. In this report, we describe its integration with automated image-processing capabilities, allowing the continuous monitoring of key cell culture characteristics. A machine learning-based algorithm enabled the specific detection of one cell type within a co-culture setting, such as human embryonic stem cells against the background of fibroblast cells. In addition, the algorithm did not confuse image artifacts resulting from microfabrication, such as scratches on surfaces, or dust particles, with cellular features. We demonstrate how the automation of flow control, environmental control, and image acquisition can be employed to image the whole culture area and obtain time-course data of mouse embryonic stem cell cultures, for example, for confluency.

Automated method for the rapid and precise estimation of adherent cell culture characteristics from phase contrast microscopy images.

The quantitative determination of key adherent cell culture characteristics such as confluency, morphology, and cell density is necessary for the evaluation of experimental outcomes and to provide a suitable basis for the establishment of robust cell culture protocols. Automated processing of images acquired using phase contrast microscopy (PCM), an imaging modality widely used for the visual inspection of adherent cell cultures, could enable the non-invasive determination of these characteristics. We present an image-processing approach that accurately detects cellular objects in PCM images through a combination of local contrast thresholding and post hoc correction of halo artifacts. The method was thoroughly validated using a variety of cell lines, microscope models and imaging conditions, demonstrating consistently high segmentation performance in all cases and very short processing times (<1 s per 1,208 × 960 pixels image). Based on the high segmentation performance, it was possible to precisely determine culture confluency, cell density, and the morphology of cellular objects, demonstrating the wide applicability of our algorithm for typical microscopy image processing pipelines. Furthermore, PCM image segmentation was used to facilitate the interpretation and analysis of fluorescence microscopy data, enabling the determination of temporal and spatial expression patterns of a fluorescent reporter. We created a software toolbox (PHANTAST) that bundles all the algorithms and provides an easy to use graphical user interface. Source-code for MATLAB and ImageJ is freely available under a permissive open-source license.

Microfabricated modular scale-down device for regenerative medicine process development.

The capacity of milli and micro litre bioreactors to accelerate process development has been successfully demonstrated in traditional biotechnology. However, for regenerative medicine present smaller scale culture methods cannot cope with the wide range of processing variables that need to be evaluated. Existing microfabricated culture devices, which could test different culture variables with a minimum amount of resources (e.g. expensive culture medium), are typically not designed with process development in mind. We present a novel, autoclavable, and microfabricated scale-down device designed for regenerative medicine process development. The microfabricated device contains a re-sealable culture chamber that facilitates use of standard culture protocols, creating a link with traditional small-scale culture devices for validation and scale-up studies. Further, the modular design can easily accommodate investigation of different culture substrate/extra-cellular matrix combinations. Inactivated mouse embryonic fibroblasts (iMEF) and human embryonic stem cell (hESC) colonies were successfully seeded on gelatine-coated tissue culture polystyrene (TC-PS) using standard static seeding protocols. The microfluidic chip included in the device offers precise and accurate control over the culture medium flow rate and resulting shear stresses in the device. Cells were cultured for two days with media perfused at 300 µl.h(-1) resulting in a modelled shear stress of 1.1×10(-4) Pa. Following perfusion, hESC colonies stained positively for different pluripotency markers and retained an undifferentiated morphology. An image processing algorithm was developed which permits quantification of co-cultured colony-forming cells from phase contrast microscope images. hESC colony sizes were quantified against the background of the feeder cells (iMEF) in less than 45 seconds for high-resolution images, which will permit real-time monitoring of culture progress in future experiments. The presented device is a first step to harness the advantages of microfluidics for regenerative medicine process development.

Microfluidic approaches for systems and synthetic biology.

Microfluidic systems miniaturise biological experimentation leading to reduced sample volume, analysis time and cost. Recent innovations have allowed the application of -omics approaches on the microfluidic scale. It is now possible to perform 1.5 million PCR reactions simultaneously, obtain transcriptomic data from as little as 150 cells (as few as 2 transcripts per gene of interest) and perform mass-spectrometric analyses online. For synthetic biology, unit operations have been developed that allow de novo construction of synthetic systems from oligonucleotide synthesis through to high-throughput, high efficiency electroporation of single cells or encapsulation into abiotic chassis enabling the processing of thousands of synthetic organisms per hour. Future directions include a push towards integrating more processes into a single device and replacing off-chip analyses where possible.

Calcium phosphate transfection generates mammalian recombinant cell lines with higher specific productivity than polyfection.

Transfection with polyethylenimine (PEI) was evaluated as a method for the generation of recombinant Chinese hamster ovary (CHO DG44) cell lines by direct comparison with calcium phosphate-DNA coprecipitation (CaPO4) using both green fluorescent protein (GFP) and a monoclonal antibody as reporter proteins. Following transfection with a GFP expression vector, the proportion of GFP-positive cells as determined by flow cytometry was fourfold higher for the PEI transfection as compared to the CaPO4 transfection. However, the mean level of transient GFP expression for the cells with the highest level of fluorescence was twofold greater for the CaPO4 transfection. Fluorescence in situ hybridization on metaphase chromosomes from pools of cells grown under selective pressure demonstrated that plasmid integration always occurred at a single site regardless of the transfection method. Importantly, the copy number of integrated plasmids was measurably higher in cells transfected with CaPO4. The efficiency of recombinant cell line recovery under selective pressure was fivefold higher following PEI transfection, but the average specific productivity of a recombinant antibody was about twofold higher for the CaPO4-derived cell lines. Nevertheless, no difference between the two transfection methods was observed in terms of the stability of protein production. These results demonstrated the feasibility of generating recombinant CHO-derived cell lines by PEI transfection. However, this method appeared inferior to CaPO4 transfection with regard to the specific productivity of the recovered cell lines.

New disposable tubes for rapid and precise biomass assessment for suspension cultures of mammalian cells.

We present a new approach for biomass assessment in cell culture using a disposable microcentrifuge tube. The specially designed tube is fitted with an upper chamber for sample loading and a lower 5 microL capillary for cell collection during centrifugation. The resulting packed cell volume (PCV) can be quantitatively expressed as the percentage of the total volume of the sample. The present study focused on the validation of the method with mammalian cell lines that are widely used in bioprocessing. Using several examples, the PCV method was shown to be more precise, rapid, and reproducible than manual cell counting.