Large Image Libraries Research Articles

Machine learning to support visual auditing of home-based lateral flow immunoassay self-test results for SARS-CoV-2 antibodies

BackgroundLateral flow immunoassays (LFIAs) are being used worldwide for COVID-19 mass testing and antibody prevalence studies. Relatively simple to use and low cost, these tests can be self-administered at home, but rely on subjective interpretation of a test line by eye, risking false positives and false negatives. Here, we report on the development of ALFA (Automated Lateral Flow Analysis) to improve reported sensitivity and specificity.MethodsOur computational pipeline uses machine learning, computer vision techniques and signal processing algorithms to analyse images of the Fortress LFIA SARS-CoV-2 antibody self-test, and subsequently classify results as invalid, IgG negative and IgG positive. A large image library of 595,339 participant-submitted test photographs was created as part of the REACT-2 community SARS-CoV-2 antibody prevalence study in England, UK. Alongside ALFA, we developed an analysis toolkit which could also detect device blood leakage issues.ResultsAutomated analysis showed substantial agreement with human experts (Cohen’s kappa 0.90–0.97) and performed consistently better than study participants, particularly for weak positive IgG results. Specificity (98.7–99.4%) and sensitivity (90.1–97.1%) were high compared with visual interpretation by human experts (ranges due to the varying prevalence of weak positive IgG tests in datasets).ConclusionsGiven the potential for LFIAs to be used at scale in the COVID-19 response (for both antibody and antigen testing), even a small improvement in the accuracy of the algorithms could impact the lives of millions of people by reducing the risk of false-positive and false-negative result read-outs by members of the public. Our findings support the use of machine learning-enabled automated reading of at-home antibody lateral flow tests as a tool for improved accuracy for population-level community surveillance.

Deep Horizon: A machine learning network that recovers accreting black hole parameters

Context. The Event Horizon Telescope recently observed the first shadow of a black hole. Images like this can potentially be used to test or constrain theories of gravity and deepen the understanding in plasma physics at event horizon scales, which requires accurate parameter estimations. Aims. In this work, we present Deep Horizon, two convolutional deep neural networks that recover the physical parameters from images of black hole shadows. We investigate the effects of a limited telescope resolution and observations at higher frequencies. Methods. We trained two convolutional deep neural networks on a large image library of simulated mock data. The first network is a Bayesian deep neural regression network and is used to recover the viewing angle i, and position angle, mass accretion rate Ṁ, electron heating prescription Rhigh and the black hole mass MBH. The second network is a classification network that recovers the black hole spin a. Results. We find that with the current resolution of the Event Horizon Telescope, it is only possible to accurately recover a limited number of parameters of a static image, namely the mass and mass accretion rate. Since potential future space-based observing missions will operate at frequencies above 230 GHz, we also investigated the applicability of our network at a frequency of 690 GHz. The expected resolution of space-based missions is higher than the current resolution of the Event Horizon Telescope, and we show that Deep Horizon can accurately recover the parameters of simulated observations with a comparable resolution to such missions.

Dictionary-enhanced imaging cytometry

State-of-the-art high-throughput microscopes are now capable of recording image data at a phenomenal rate, imaging entire microscope slides in minutes. In this paper we investigate how a large image set can be used to perform automated cell classification and denoising. To this end, we acquire an image library consisting of over one quarter-million white blood cell (WBC) nuclei together with CD15/CD16 protein expression for each cell. We show that the WBC nucleus images alone can be used to replicate CD expression-based gating, even in the presence of significant imaging noise. We also demonstrate that accurate estimates of white blood cell images can be recovered from extremely noisy images by comparing with a reference dictionary. This has implications for dose-limited imaging when samples belong to a highly restricted class such as a well-studied cell type. Furthermore, large image libraries may endow microscopes with capabilities beyond their hardware specifications in terms of sensitivity and resolution. We call for researchers to crowd source large image libraries of common cell lines to explore this possibility.

AccessEmergency Medicine: A Review

AccessEmergency Medicine from McGraw-Hill is designed to meet the needs of emergency departments with quick information to aid in diagnosing and treating Emergency Department (ED) complaints. Students, clinical instructors, and residents will also find the resource's self-study components and textbooks useful. The resource also contains both live and animated videos of common procedures, cases, a large image library, drug information, and patient education materials.

Development of a new grading scale for tear ferning.

This paper reports on the development of a new tear ferning (TF) subjective grading scale, and compares it with the Rolando scale. TF patterns obtained from tear film samples collected from normal and dry eye subjects in previous studies were collated into a large image library. From this library, 60 images were selected to represent the full range of possible TF patterns, and a further sub-set of 15 images was chosen for analysis. Twenty-five optometrists were asked to rank the images in increasing order between extreme anchors on a scale of TF patterns. Interim statistical analysis of this ranking found 7 homogeneous sub-sets, where the image rankings overlapped for a group of images. A representative image (typically the mean) from each group was then adopted as the grade standard. Using this new 7-point grading scale, 25 optometrists were asked to grade the entire 60 image library at two sessions: once using the 4-point Rolando scale and once using the new 7-point scale, applying 0.25 grade unit interpolation. Statistical analysis found that for the larger image set, the Rolando scale produced 3 homogeneous sub-sets, and the 7-point scale produced 5 homogeneous sub-sets. With this refinement, a new 5-point TF scale (Grades 0-4) was obtained. The Rolando grading scale lacks discrimination between its Type I and II grades, reducing its reliability. The new 5-point grading scale is able to differentiate between TF patterns, and may provide additional support for the use of TF for both researcher and clinician.

Shape recognition from large image libraries by inexact graph matching

This paper describes a graph-matching technique for recognising line-pattern shapes in large image databases. The methodological contribution of the paper is to develop a Bayesian matching algorithm that uses edge-consistency and node attribute similarity. This information is used to determine the a posteriori probability of a query graph for each of the candidate matches in the database. The node feature-vectors are constructed by computing normalised histograms of pairwise geometric attributes. Attribute similarity is assessed by computing the Bhattacharyya distance between the histograms. Recognition is realised by selecting the candidate from the database which has the largest a posteriori probability.

Image libraries and the Internet

Digital imaging is becoming a major information resource within the Internet. Although image access tools are primitive today, they are evolving rapidly. We look at content descriptor-based and pixel-based image queries for accessing large image libraries. The access methods will be combined and enhanced in future retrieval systems.