Leveraging deep learning for detecting red blood cell morphological changes in blood films from children with severe malaria anaemia.

In sub-Saharan Africa, acute-onset severe malaria anaemia (SMA) is a critical challenge, particularly affecting children under five. The acute drop in haematocrit in SMA is thought to be driven by an increased phagocytotic pathological process in the spleen, leading to the presence of distinct red blood cells (RBCs) with altered morphological characteristics. We hypothesized that these RBCs could be detected systematically and at scale in peripheral blood films (PBFs) by harnessing the capabilities of deep learning models. Assessment of PBFs by a microscopist does not scale for this task and is subject to variability. Here we introduce a deep learning model, leveraging a weakly supervised Multiple Instance Learning framework, to Identify SMA (MILISMA) through the presence of morphologically changed RBCs. MILISMA achieved a classification accuracy of 83% (receiver operating characteristic area under the curve [AUC] of 87%; precision-recall AUC of 76%). More importantly, MILISMA's capabilities extend to identifying statistically significant morphological distinctions (p < 0.01) in RBCs descriptors. Our findings are enriched by visual analyses, which underscore the unique morphological features of SMA-affected RBCs when compared to non-SMA cells. This model aided detection and characterization of RBC alterations could enhance the understanding of SMA's pathology and refine SMA diagnostic and prognostic evaluation processes at scale.

Optical mesoscopy, machine learning, and computational microscopy enable high information content diagnostic imaging of blood films.

Automated image-based assessment of blood films has tremendous potential to support clinical haematology within overstretched healthcare systems. To achieve this, efficient and reliable digital capture of the rich diagnostic information contained within a blood film is a critical first step. However, this is often challenging, and in many cases entirely unfeasible, with the microscopes typically used in haematology due to the fundamental trade-off between magnification and spatial resolution. To address this, we investigated three state-of-the-art approaches to microscopic imaging of blood films which leverage recent advances in optical and computational imaging and analysis to increase the information capture capacity of the optical microscope: optical mesoscopy, which uses a giant microscope objective (Mesolens) to enable high-resolution imaging at low magnification; Fourier ptychographic microscopy, a computational imaging method which relies on oblique illumination with a series of LEDs to capture high-resolution information; and deep neural networks which can be trained to increase the quality of low magnification, low resolution images. We compare and contrast the performance of these techniques for blood film imaging for the exemplar case of Giemsa-stained peripheral blood smears. Using computational image analysis and shape-based object classification, we demonstrate their use for automated analysis of red blood cell morphology and visualization and detection of small blood-borne parasites such as the malarial parasite Plasmodium falciparum. Our results demonstrate that these new methods greatly increase the information capturing capacity of the light microscope, with transformative potential for haematology and more generally across digital pathology. © 2021 The Authors. The Journal of Pathology published by John Wiley & Sons, Ltd. on behalf of The Pathological Society of Great Britain and Ireland.

Structure-dependent amplification for denoising and background correction in Fourier ptychographic microscopy.

Fourier Ptychographic Microscopy (FPM) allows high resolution imaging using iterative phase retrieval to recover an estimate of the complex object from a series of images captured under oblique illumination. FPM is particularly sensitive to noise and uncorrected background signals as it relies on combining information from brightfield and noisy darkfield (DF) images. In this article we consider the impact of different noise sources in FPM and show that inadequate removal of the DF background signal and associated noise are the predominant cause of artefacts in reconstructed images. We propose a simple solution to FPM background correction and denoising that outperforms existing methods in terms of image quality, speed and simplicity, whilst maintaining high spatial resolution and sharpness of the reconstructed image. Our method takes advantage of the data redundancy in real space within the acquired dataset to boost the signal-to-background ratio in the captured DF images, before optimally suppressing background signal. By incorporating differentially denoised images within the classic FPM iterative phase retrieval algorithm, we show that it is possible to achieve efficient removal of background artefacts without suppression of high frequency information. The method is tested using simulated data and experimental images of thin blood films, bone marrow and liver tissue sections. Our approach is non-parametric, requires no prior knowledge of the noise distribution and can be directly applied to other hardware platforms and reconstruction algorithms making it widely applicable in FPM.

Depleted circulatory complement-lysis inhibitor (CLI) in childhood cerebral malaria returns to normal with convalescence.

BACKGROUND: Cerebral malaria (CM), is a life-threatening childhood malaria syndrome with high mortality. CM is associated with impaired consciousness and neurological damage. It is not fully understood, as yet, why some children develop CM. Presented here is an observation from longitudinal studies on CM in a paediatric cohort of children from a large, densely-populated and malaria holoendemic, sub-Saharan, West African metropolis. METHODS: Plasma samples were collected from a cohort of children with CM, severe malarial anaemia (SMA), uncomplicated malaria (UM), non-malaria positive healthy community controls (CC), and coma and anemic patients without malaria, as disease controls (DC). Proteomic two-dimensional difference gel electrophoresis (2D-DIGE) and mass spectrometry were used in a discovery cohort to identify plasma proteins that might be discriminatory among these clinical groups. The circulatory levels of identified proteins of interest were quantified by ELISA in a prospective validation cohort. RESULTS: The proteome analysis revealed differential abundance of circulatory complement-lysis inhibitor (CLI), also known as Clusterin (CLU). CLI circulatory level was low at hospital admission in all children presenting with CM and recovered to normal level during convalescence (p < 0.0001). At acute onset, circulatory level of CLI in the CM group significantly discriminates CM from the UM, SMA, DC and CC groups. CONCLUSIONS: The CLI circulatory level is low in all patients in the CM group at admission, but recovers through convalescence. The level of CLI at acute onset may be a specific discriminatory marker of CM. This work suggests that CLI may play a role in the pathophysiology of CM and may be useful in the diagnosis and follow-up of children presenting with CM.

Expert-level automated malaria diagnosis on routine blood films with deep neural networks.

Over 200 million malaria cases globally lead to half a million deaths annually. Accurate malaria diagnosis remains a challenge. Automated imaging processing approaches to analyze Thick Blood Films (TBF) could provide scalable solutions, for urban healthcare providers in the holoendemic malaria sub-Saharan region. Although several approaches have been attempted to identify malaria parasites in TBF, none have achieved negative and positive predictive performance suitable for clinical use in the west sub-Saharan region. While malaria parasite object detection remains an intermediary step in achieving automatic patient diagnosis, training state-of-the-art deep-learning object detectors requires the human-expert labor-intensive process of labeling a large dataset of digitized TBF. To overcome these challenges and to achieve a clinically usable system, we show a novel approach. It leverages routine clinical-microscopy labels from our quality-controlled malaria clinics, to train a Deep Malaria Convolutional Neural Network classifier (DeepMCNN) for automated malaria diagnosis. Our system also provides total Malaria Parasite (MP) and White Blood Cell (WBC) counts allowing parasitemia estimation in MP/µL, as recommended by the WHO. Prospective validation of the DeepMCNN achieves sensitivity/specificity of 0.92/0.90 against expert-level malaria diagnosis. Our approach PPV/NPV performance is of 0.92/0.90, which is clinically usable in our holoendemic settings in the densely populated metropolis of Ibadan. It is located within the most populous African country (Nigeria) and with one of the largest burdens of Plasmodium falciparum malaria. Our openly available method is of importance for strategies aimed to scale malaria diagnosis in urban regions where daily assessment of thousands of specimens is required.

The IL17F and IL17RA Genetic Variants Increase Risk of Cerebral Malaria in Two African Populations.

Cerebral malaria (CM) is a neurological complication of infection with Plasmodium falciparum that is partly caused by cytokine-mediated inflammation. It is not known whether interleukin-17 (IL-17) cytokines, which regulate inflammation, control the development of CM. To evaluate the involvement of IL-17 cytokines in CM, we analyzed 46 common polymorphisms in IL17A, IL17F, and IL17RA (which encodes the common receptor chain of the members of the IL-17 family) in two independent African populations. A case-control study involving 115 Nigerian children with CM and 160 controls from the community (CC) showed that IL17F reference single nucleotide polymorphism (SNP) 6913472 (rs6913472) (P = 0.004; odds ratio [OR] = 3.12), IL17F rs4715291 (P = 0.004; OR = 2.82), IL17RA rs12159217 (P = 0.01; OR = 2.27), and IL17RA rs41396547 (P = 0.026; OR = 3.15) were independently associated with CM. A replication study was performed in 240 nuclear Malian family trios (two parents with one CM child). We replicated the association for 3 SNPs, IL17F rs6913472 (P = 0.03; OR = 1.39), IL17RA rs12159217 (P = 0.01; OR = 1.52), and IL17RA rs41396547 (P = 0.04; OR = 3.50). We also found that one additional SNP, IL17RA rs41433045, in linkage disequilibrium (LD) with rs41396547, was associated with CM in both Nigeria and Mali (P = 0.002; OR = 4.12 in the combined sample). We excluded the possibility that SNPs outside IL17F and IL17RA, in strong LD with the associated SNPs, could account for the observed associations. Furthermore, the results of a functional study indicated that the aggravating GA genotype of IL17F rs6913472 was associated with lower IL-17F concentrations. Our findings show for the first time that IL17F and IL17RA polymorphisms modulate susceptibility to CM and provide evidence that IL-17F protects against CM.

Profiling persistent tubercule bacilli from patient sputa during therapy predicts early drug efficacy.

BACKGROUND: New treatment options are needed to maintain and improve therapy for tuberculosis, which caused the death of 1.5 million people in 2013 despite potential for an 86 % treatment success rate. A greater understanding of Mycobacterium tuberculosis (M.tb) bacilli that persist through drug therapy will aid drug development programs. Predictive biomarkers for treatment efficacy are also a research priority. METHODS AND RESULTS: Genome-wide transcriptional profiling was used to map the mRNA signatures of M.tb from the sputa of 15 patients before and 3, 7 and 14 days after the start of standard regimen drug treatment. The mRNA profiles of bacilli through the first 2 weeks of therapy reflected drug activity at 3 days with transcriptional signatures at days 7 and 14 consistent with reduced M.tb metabolic activity similar to the profile of pre-chemotherapy bacilli. These results suggest that a pre-existing drug-tolerant M.tb population dominates sputum before and after early drug treatment, and that the mRNA signature at day 3 marks the killing of a drug-sensitive sub-population of bacilli. Modelling patient indices of disease severity with bacterial gene expression patterns demonstrated that both microbiological and clinical parameters were reflected in the divergent M.tb responses and provided evidence that factors such as bacterial load and disease pathology influence the host-pathogen interplay and the phenotypic state of bacilli. Transcriptional signatures were also defined that predicted measures of early treatment success (rate of decline in bacterial load over 3 days, TB test positivity at 2 months, and bacterial load at 2 months). CONCLUSIONS: This study defines the transcriptional signature of M.tb bacilli that have been expectorated in sputum after two weeks of drug therapy, characterizing the phenotypic state of bacilli that persist through treatment. We demonstrate that variability in clinical manifestations of disease are detectable in bacterial sputa signatures, and that the changing M.tb mRNA profiles 0-2 weeks into chemotherapy predict the efficacy of treatment 6 weeks later. These observations advocate assaying dynamic bacterial phenotypes through drug therapy as biomarkers for treatment success.

Affinity proteomics reveals elevated muscle proteins in plasma of children with cerebral malaria.

Systemic inflammation and sequestration of parasitized erythrocytes are central processes in the pathophysiology of severe Plasmodium falciparum childhood malaria. However, it is still not understood why some children are more at risks to develop malaria complications than others. To identify human proteins in plasma related to childhood malaria syndromes, multiplex antibody suspension bead arrays were employed. Out of the 1,015 proteins analyzed in plasma from more than 700 children, 41 differed between malaria infected children and community controls, whereas 13 discriminated uncomplicated malaria from severe malaria syndromes. Markers of oxidative stress were found related to severe malaria anemia while markers of endothelial activation, platelet adhesion and muscular damage were identified in relation to children with cerebral malaria. These findings suggest the presence of generalized vascular inflammation, vascular wall modulations, activation of endothelium and unbalanced glucose metabolism in severe malaria. The increased levels of specific muscle proteins in plasma implicate potential muscle damage and microvasculature lesions during the course of cerebral malaria.

Circulatory hepcidin is associated with the anti-inflammatory response but not with iron or anemic status in childhood malaria.

Cerebral malaria (CM) and severe malarial anemia (SMA) are the most serious life-threatening clinical syndromes of Plasmodium falciparum infection in childhood. Therefore, it is important to understand the pathology underlying the development of CM and SMA as opposed to uncomplicated malaria (UM). Increased levels of hepcidin have been associated with UM, but its level and role in severe malarial disease remains to be investigated. Plasma and clinical data were obtained as part of a prospective case-control study of severe childhood malaria at the main tertiary hospital of the city of Ibadan, Nigeria. Here, we report that hepcidin levels are lower in children with SMA or CM than in those with milder outcome (UM). While different profiles of pro- and anti-inflammatory cytokines were observed between the malaria syndromes, circulatory hepcidin levels remained associated with the levels of its regulatory cytokine interleukin-6 and of the anti-inflammatory cytokine inerleukin-10, irrespective of iron status, anemic status, and general acute-phase response. We propose a role for hepcidin in anti-inflammatory processes in childhood malaria.

Biomarker discovery by sparse canonical correlation analysis of complex clinical phenotypes of tuberculosis and malaria.

Biomarker discovery aims to find small subsets of relevant variables in 'omics data that correlate with the clinical syndromes of interest. Despite the fact that clinical phenotypes are usually characterized by a complex set of clinical parameters, current computational approaches assume univariate targets, e.g. diagnostic classes, against which associations are sought for. We propose an approach based on asymmetrical sparse canonical correlation analysis (SCCA) that finds multivariate correlations between the 'omics measurements and the complex clinical phenotypes. We correlated plasma proteomics data to multivariate overlapping complex clinical phenotypes from tuberculosis and malaria datasets. We discovered relevant 'omic biomarkers that have a high correlation to profiles of clinical measurements and are remarkably sparse, containing 1.5-3% of all 'omic variables. We show that using clinical view projections we obtain remarkable improvements in diagnostic class prediction, up to 11% in tuberculosis and up to 5% in malaria. Our approach finds proteomic-biomarkers that correlate with complex combinations of clinical-biomarkers. Using the clinical-biomarkers improves the accuracy of diagnostic class prediction while not requiring the measurement plasma proteomic profiles of each subject. Our approach makes it feasible to use omics' data to build accurate diagnostic algorithms that can be deployed to community health centres lacking the expensive 'omics measurement capabilities.

A one-step method for generating antimicrobial nanofibre meshes via coaxial electrospinning.

Respiratory diseases, including influenza, infectious pneumonia, and severe acute respiratory syndrome (SARS), are a leading cause of morbidity and mortality worldwide. The recent COVID-19 pandemic claimed over 6.9 million lives globally. With the possibility of future pandemics, the creation of affordable antimicrobial meshes for protective gear, such as facemasks, is essential. Electrospinning has been a focus for much of this research, but most approaches are complex and expensive, often wasting raw materials by distributing antiviral agents throughout the mesh despite the fact they can only be active if at the fibre surface. Here, we report a low cost and efficient one-step method to produce nanofibre meshes with antimicrobial activity, including against SARS-CoV-2. Cetrimonium bromide (CTAB) was deposited directly onto the surface of polycaprolactone (PCL) fibres by coaxial electrospinning. The CTAB-coated samples have denser meshes with finer nanofibres than non-coated PCL fibres (mean diameter: ∼300 nm versus ∼900 nm, with mean pore size: ∼300 nm versus > 600 nm). The formulations have > 90% coating efficiency and exhibit a burst release of CTAB upon coming into contact with aqueous media. The CTAB-coated materials have strong antibacterial activity against Staphylococcus aureus (ca. 100%) and Pseudomonas aeruginosa (96.5 ± 4.1%) bacteria, as well as potent antiviral activity with over 99.9% efficacy against both respiratory syncytial virus and SARS-CoV-2. The CTAB-coated nanofibre mesh thus has great potential to form a mask material for preventing both bacterial and viral respiratory infections.

Specialist hybrid models with asymmetric training for malaria prevalence prediction.

Malaria is a common and serious disease that primarily affects developing countries and its spread is influenced by a variety of environmental and human behavioral factors; therefore, accurate prevalence prediction has been identified as a critical component of the Global Technical Strategy for Malaria from 2016 to 2030. While traditional differential equation models can perform basic forecasting, supervised machine learning algorithms provide more accurate predictions, as demonstrated by a recent study using an elastic net model (REMPS). Nevertheless, current short-term prediction systems do not achieve the required accuracy levels for routine clinical practice. To improve in this direction, stacked hybrid models have been proposed, in which the outputs of several machine learning models are aggregated by using a meta-learner predictive model. In this paper, we propose an alternative specialist hybrid approach that combines a linear predictive model that specializes in the linear component of the malaria prevalence signal and a recurrent neural network predictive model that specializes in the non-linear residuals of the linear prediction, trained with a novel asymmetric loss. Our findings show that the specialist hybrid approach outperforms the current state-of-the-art stacked models on an open-source dataset containing 22 years of malaria prevalence data from the city of Ibadan in southwest Nigeria. The specialist hybrid approach is a promising alternative to current prediction methods, as well as a tool to improve decision-making and resource allocation for malaria control in high-risk countries.

Detection of acute promyelocytic leukemia in peripheral blood and bone marrow with annotation-free deep learning.

While optical microscopy inspection of blood films and bone marrow aspirates by a hematologist is a crucial step in establishing diagnosis of acute leukemia, especially in low-resource settings where other diagnostic modalities are not available, the task remains time-consuming and prone to human inconsistencies. This has an impact especially in cases of Acute Promyelocytic Leukemia (APL) that require urgent treatment. Integration of automated computational hematopathology into clinical workflows can improve the throughput of these services and reduce cognitive human error. However, a major bottleneck in deploying such systems is a lack of sufficient cell morphological object-labels annotations to train deep learning models. We overcome this by leveraging patient diagnostic labels to train weakly-supervised models that detect different types of acute leukemia. We introduce a deep learning approach, Multiple Instance Learning for Leukocyte Identification (MILLIE), able to perform automated reliable analysis of blood films with minimal supervision. Without being trained to classify individual cells, MILLIE differentiates between acute lymphoblastic and myeloblastic leukemia in blood films. More importantly, MILLIE detects APL in blood films (AUC 0.94 ± 0.04) and in bone marrow aspirates (AUC 0.99 ± 0.01). MILLIE is a viable solution to augment the throughput of clinical pathways that require assessment of blood film microscopy.

Low-field magnetic resonance image enhancement via stochastic image quality transfer.

Low-field (<1T) magnetic resonance imaging (MRI) scanners remain in widespread use in low- and middle-income countries (LMICs) and are commonly used for some applications in higher income countries e.g. for small child patients with obesity, claustrophobia, implants, or tattoos. However, low-field MR images commonly have lower resolution and poorer contrast than images from high field (1.5T, 3T, and above). Here, we present Image Quality Transfer (IQT) to enhance low-field structural MRI by estimating from a low-field image the image we would have obtained from the same subject at high field. Our approach uses (i) a stochastic low-field image simulator as the forward model to capture uncertainty and variation in the contrast of low-field images corresponding to a particular high-field image, and (ii) an anisotropic U-Net variant specifically designed for the IQT inverse problem. We evaluate the proposed algorithm both in simulation and using multi-contrast (T1-weighted, T2-weighted, and fluid attenuated inversion recovery (FLAIR)) clinical low-field MRI data from an LMIC hospital. We show the efficacy of IQT in improving contrast and resolution of low-field MR images. We demonstrate that IQT-enhanced images have potential for enhancing visualisation of anatomical structures and pathological lesions of clinical relevance from the perspective of radiologists. IQT is proved to have capability of boosting the diagnostic value of low-field MRI, especially in low-resource settings.

Global transcriptional analysis of primitive thymocytes reveals accelerated dynamics of T cell specification in fetal stages.

T cell development constitutes a multistage process allowing the dissection of events resulting in cellular commitment and functional specification in a specialized microenvironment. This process is guided by the appropriate expression of regulatory genetic factors like transcriptional activators or repressors which are, in part, dependent on instructive signals of the microenvironment. To date, it remains unclear whether exactly the same genetic mechanism acts in adult compared to fetal T cell development. In order to directly compare T cell commitment during adult and fetal differentiation, we isolated subsequent stages of intrathymic subpopulations starting with early canonical T cell progenitors up to irreversibly committed T cell precursors. The genome-wide analysis revealed several distinct gene clusters with a specific pattern of gene regulation for each subset. The largest cluster contained genes upregulated after transition through the most primitive pool into the next transitory population with a consistently elevated expression of elements associated with ongoing T cell fate specification, like Gata3 and Tcf7, in fetal progenitors. Furthermore, adult and fetal T cell progenitors occupied distinct "transcriptional territories" revealing a precise land map of the progression to final T cell commitment operating in different developmental windows. The presence and/or elevated expression of elements associated with an ongoing establishment of a T cell signature in the most primitive fetal subset is highly suggestive for an extrathymic initiation of T cell specification and underlines the fundamental differences in fetal versus adult lymphopoiesis.

Haemoglobinuria among children with severe malaria attending tertiary care in Ibadan, Nigeria.

BACKGROUND: Haemoglobinuria is one of the manifestations of severe malaria and results from severe intravascular haemolysis. Glucose-6-phosphate dehydrogenase (G6PD) deficiency has been implicated in its aetiology. Haemoglobinuria may be associated with severe anaemia and, less frequently, acute renal failure. METHODS: A prospective case-control study was carried out to determine the incidence of haemoglobinuria as confirmed by dipstick urinalysis, microscopy and spectrophotometric measurement, among children with severe malaria. A total of 251 children presenting at the Children's Emergency Ward with severe malaria were recruited over a period of 21 months. The G6PD status and the outcomes of severe malaria in children with and without haemoglobinuria was studied with respect to renal failure, the recurrence of haemoglobinuria and blood pressure changes over a three-month follow-up period. RESULTS: It was found that the incidence of haemoglobinuria among children with severe malaria is 19.1%. Children <5 years constituted 76.8% of all the study patients. Patients with haemoglobinuria had median age of 52.5 months, which was significantly higher than 35 months in patients without haemoglobinuria (p=0.001). Although, haemaglobinuria was commoner among boys (54.2%) than girls (45.8%), the difference was not statistically significant. There were no significant differences between children with and without haemoglobinuria regarding their nutritional status or parasite densities. Among the clinical features of the study patients, only jaundice was significantly associated with haemoglobinuria (p=0.0001). Renal failure occurred in three out of 48 children with haemoglobinuria and in none of the 203 without. There was not recurrence of haemoglobinuria in the follow-up period. At discharge, blood pressure was elevated in six children (one previously haemoglobinuric), but all returned to normal within the follow-up period. CONCLUSIONS: Haemoglobinuria was a prominent feature of severe malaria and it was significantly associated with jaundice at presentation. Haemoglobinuria was commoner in older children than younger children but not related to sex. G6PD deficiency was not an independent predictor of the occurrence or outcome of haemoglobinuria. Blood pressure was not affected by haemoglobinuria on admission nor during follow-up.

Network topological determinants of pathogen spread.

How do we best constrain social interactions to decrease transmission of communicable diseases? Indiscriminate suppression is unsustainable long term and presupposes that all interactions carry equal importance. Instead, transmission within a social network has been shown to be determined by its topology. In this paper, we deploy simulations to understand and quantify the impact on disease transmission of a set of topological network features, building a dataset of 9000 interaction graphs using generators of different types of synthetic social networks. Independently of the topology of the network, we maintain constant the total volume of social interactions in our simulations, to show how even with the same social contact some network structures are more or less resilient to the spread. We find a suitable intervention to be specific suppression of unfamiliar and casual interactions that contribute to the network's global efficiency. This is, pathogen spread is significantly reduced by limiting specific kinds of contact rather than their global number. Our numerical studies might inspire further investigation in connection to public health, as an integrative framework to craft and evaluate social interventions in communicable diseases with different social graphs or as a highlight of network metrics that should be captured in social studies.

Content aware multi-focus image fusion for high-magnification blood film microscopy.

Automated digital high-magnification optical microscopy is key to accelerating biology research and improving pathology clinical pathways. High magnification objectives with large numerical apertures are usually preferred to resolve the fine structural details of biological samples, but they have a very limited depth-of-field. Depending on the thickness of the sample, analysis of specimens typically requires the acquisition of multiple images at different focal planes for each field-of-view, followed by the fusion of these planes into an extended depth-of-field image. This translates into low scanning speeds, increased storage space, and processing time not suitable for high-throughput clinical use. We introduce a novel content-aware multi-focus image fusion approach based on deep learning which extends the depth-of-field of high magnification objectives effectively. We demonstrate the method with three examples, showing that highly accurate, detailed, extended depth of field images can be obtained at a lower axial sampling rate, using 2-fold fewer focal planes than normally required.

Representational ethical model calibration.

Equity is widely held to be fundamental to the ethics of healthcare. In the context of clinical decision-making, it rests on the comparative fidelity of the intelligence - evidence-based or intuitive - guiding the management of each individual patient. Though brought to recent attention by the individuating power of contemporary machine learning, such epistemic equity arises in the context of any decision guidance, whether traditional or innovative. Yet no general framework for its quantification, let alone assurance, currently exists. Here we formulate epistemic equity in terms of model fidelity evaluated over learnt multidimensional representations of identity crafted to maximise the captured diversity of the population, introducing a comprehensive framework for Representational Ethical Model Calibration. We demonstrate the use of the framework on large-scale multimodal data from UK Biobank to derive diverse representations of the population, quantify model performance, and institute responsive remediation. We offer our approach as a principled solution to quantifying and assuring epistemic equity in healthcare, with applications across the research, clinical, and regulatory domains.

Development, deployment and evaluation of digitally enabled, remote, supported rehabilitation for people with long COVID-19 (Living With COVID-19 Recovery): protocol for a mixed-methods study.

INTRODUCTION: Long COVID-19 is a distressing, disabling and heterogeneous syndrome often causing severe functional impairment. Predominant symptoms include fatigue, cognitive impairment ('brain fog'), breathlessness and anxiety or depression. These symptoms are amenable to rehabilitation delivered by skilled healthcare professionals, but COVID-19 has put severe strain on healthcare systems. This study aims to explore whether digitally enabled, remotely supported rehabilitation for people with long COVID-19 can enable healthcare systems to provide high quality care to large numbers of patients within the available resources. Specific objectives are to (1) develop and refine a digital health intervention (DHI) that supports patient assessment, monitoring and remote rehabilitation; (2) develop implementation models that support sustainable deployment at scale; (3) evaluate the impact of the DHI on recovery trajectories and (4) identify and mitigate health inequalities due to the digital divide. METHODS AND ANALYSIS: Mixed-methods, theoretically informed, single-arm prospective study, combining methods drawn from engineering/computer science with those from biomedicine. There are four work packages (WP), one for each objective. WP1 focuses on identifying user requirements and iteratively developing the intervention to meet them; WP2 combines qualitative data from users with learning from implementation science and normalisation process theory, to promote adoption, scale-up, spread and sustainability of the intervention; WP3 uses quantitative demographic, clinical and resource use data collected by the DHI to determine illness trajectories and how these are affected by use of the DHI; while WP4 focuses on identifying and mitigating health inequalities and overarches the other three WPs. ETHICS AND DISSEMINATION: Ethical approval obtained from East Midlands - Derby Research Ethics Committee (reference 288199). Our dissemination strategy targets three audiences: (1) Policy makers, Health service managers and clinicians responsible for delivering long COVID-19 services; (2) patients and the public; (3) academics. TRIAL REGISTRATION NUMBER: Research Registry number: researchregistry6173.