Search results
Found 8733 matches for
Distinct patterns of vital sign and inflammatory marker responses in adults with suspected bloodstream infection.
ObjectivesTo identify patterns in inflammatory marker and vital sign responses in adult with suspected bloodstream infection (BSI) and define expected trends in normal recovery.MethodsWe included patients ≥16y from Oxford University Hospitals with a blood culture taken between 01-January-2016 to 28-June-2021. We used linear and latent class mixed models to estimate trajectories in C-reactive protein (CRP), white blood count, heart rate, respiratory rate and temperature and identify CRP response subgroups. Centile charts for expected CRP responses were constructed via the lambda-mu-sigma method.ResultsIn 88,348 suspected BSI episodes; 6,908(7.8%) were culture-positive with a probable pathogen, 4,309(4.9%) contained potential contaminants, and 77,131(87.3%) were culture-negative. CRP levels generally peaked 1-2 days after blood culture collection, with varying responses for different pathogens and infection sources (p<0.0001). We identified five CRP trajectory subgroups: peak on day-1 (36,091;46.3%) or 2 (4,529;5.8%), slow recovery (10,666;13.7%), peak on day-6 (743;1.0%), and low response (25,928;33.3%). Centile reference charts tracking normal responses were constructed from those peaking on day-1/2.ConclusionsCRP and other infection response markers rise and recover differently depending on clinical syndrome and pathogen involved. However, centile reference charts, that account for these differences, can be used to track if patients are recovering line as expected and to help personalise infection.
Feasibility and usability of remote monitoring in Alzheimer's disease.
IntroductionRemote monitoring technologies (RMTs) can measure cognitive and functional decline objectively at-home, and offer opportunities to measure passively and continuously, possibly improving sensitivity and reducing participant burden in clinical trials. However, there is skepticism that age and cognitive or functional impairment may render participants unable or unwilling to comply with complex RMT protocols. We therefore assessed the feasibility and usability of a complex RMT protocol in all syndromic stages of Alzheimer's disease and in healthy control participants.MethodsFor 8 weeks, participants (N = 229) used two activity trackers, two interactive apps with either daily or weekly cognitive tasks, and optionally a wearable camera. A subset of participants participated in a 4-week sub-study (N = 45) using fixed at-home sensors, a wearable EEG sleep headband and a driving performance device. Feasibility was assessed by evaluating compliance and drop-out rates. Usability was assessed by problem rates (e.g., understanding instructions, discomfort, forgetting to use the RMT or technical problems) as discussed during bi-weekly semi-structured interviews.ResultsMost problems were found for the active apps and EEG sleep headband. Problem rates increased and compliance rates decreased with disease severity, but the study remained feasible.ConclusionsThis study shows that a highly complex RMT protocol is feasible, even in a mild-to-moderate AD population, encouraging other researchers to use RMTs in their study designs. We recommend evaluating the design of individual devices carefully before finalizing study protocols, considering RMTs which allow for real-time compliance monitoring, and engaging the partners of study participants in the research.
How Does the Proportion of Never Treatment Influence the Success of Mass Drug Administration Programs for the Elimination of Lymphatic Filariasis?
BackgroundMass drug administration (MDA) is the cornerstone for the elimination of lymphatic filariasis (LF). The proportion of the population that is never treated (NT) is a crucial determinant of whether this goal is achieved within reasonable time frames.MethodsUsing 2 individual-based stochastic LF transmission models, we assess the maximum permissible level of NT for which the 1% microfilaremia (mf) prevalence threshold can be achieved (with 90% probability) within 10 years under different scenarios of annual MDA coverage, drug combination and transmission setting.ResultsFor Anopheles-transmission settings, we find that treating 80% of the eligible population annually with ivermectin + albendazole (IA) can achieve the 1% mf prevalence threshold within 10 years of annual treatment when baseline mf prevalence is 10%, as long as NT <10%. Higher proportions of NT are acceptable when more efficacious treatment regimens are used. For Culex-transmission settings with a low (5%) baseline mf prevalence and diethylcarbamazine + albendazole (DA) or ivermectin + diethylcarbamazine + albendazole (IDA) treatment, elimination can be reached if treatment coverage among eligibles is 80% or higher. For 10% baseline mf prevalence, the target can be achieved when the annual coverage is 80% and NT ≤15%. Higher infection prevalence or levels of NT would make achieving the target more difficult.ConclusionsThe proportion of people never treated in MDA programmes for LF can strongly influence the achievement of elimination and the impact of NT is greater in high transmission areas. This study provides a starting point for further development of criteria for the evaluation of NT.
Reducing the Antigen Prevalence Target Threshold for Stopping and Restarting Mass Drug Administration for Lymphatic Filariasis Elimination: A Model-Based Cost-effectiveness Simulation in Tanzania, India and Haiti.
BackgroundThe Global Programme to Eliminate Lymphatic Filariasis (GPELF) aims to reduce and maintain infection levels through mass drug administration (MDA), but there is evidence of ongoing transmission after MDA in areas where Culex mosquitoes are the main transmission vector, suggesting that a more stringent criterion is required for MDA decision making in these settings.MethodsWe use a transmission model to investigate how a lower prevalence threshold (<1% antigenemia [Ag] prevalence compared with <2% Ag prevalence) for MDA decision making would affect the probability of local elimination, health outcomes, the number of MDA rounds, including restarts, and program costs associated with MDA and surveys across different scenarios. To determine the cost-effectiveness of switching to a lower threshold, we simulated 65% and 80% MDA coverage of the total population for different willingness to pay per disability-adjusted life-year averted for India ($446.07), Tanzania ($389.83), and Haiti ($219.84).ResultsOur results suggest that with a lower Ag threshold, there is a small proportion of simulations where extra rounds are required to reach the target, but this also reduces the need to restart MDA later in the program. For 80% coverage, the lower threshold is cost-effective across all baseline prevalences for India, Tanzania, and Haiti. For 65% MDA coverage, the lower threshold is not cost-effective due to additional MDA rounds, although it increases the probability of local elimination. Valuing the benefits of elimination to align with the GPELF goals, we find that a willingness to pay per capita government expenditure of approximately $1000-$4000 for 1% increase in the probability of local elimination would be required to make a lower threshold cost-effective.ConclusionsLower Ag thresholds for stopping MDAs generally mean a higher probability of local elimination, reducing long-term costs and health impacts. However, they may also lead to an increased number of MDA rounds required to reach the lower threshold and, therefore, increased short-term costs. Collectively, our analyses highlight that lower target Ag thresholds have the potential to assist programs in achieving lymphatic filariasis goals.
Accelerating Progress Towards the 2030 Neglected Tropical Diseases Targets: How Can Quantitative Modeling Support Programmatic Decisions?
Over the past decade, considerable progress has been made in the control, elimination, and eradication of neglected tropical diseases (NTDs). Despite these advances, most NTD programs have recently experienced important setbacks; for example, NTD interventions were some of the most frequently and severely impacted by service disruptions due to the coronavirus disease 2019 (COVID-19) pandemic. Mathematical modeling can help inform selection of interventions to meet the targets set out in the NTD road map 2021-2030, and such studies should prioritize questions that are relevant for decision-makers, especially those designing, implementing, and evaluating national and subnational programs. In September 2022, the World Health Organization hosted a stakeholder meeting to identify such priority modeling questions across a range of NTDs and to consider how modeling could inform local decision making. Here, we summarize the outputs of the meeting, highlight common themes in the questions being asked, and discuss how quantitative modeling can support programmatic decisions that may accelerate progress towards the 2030 targets.
Self-supervised learning for human activity recognition using 700,000 person-days of wearable data.
Accurate physical activity monitoring is essential to understand the impact of physical activity on one's physical health and overall well-being. However, advances in human activity recognition algorithms have been constrained by the limited availability of large labelled datasets. This study aims to leverage recent advances in self-supervised learning to exploit the large-scale UK Biobank accelerometer dataset-a 700,000 person-days unlabelled dataset-in order to build models with vastly improved generalisability and accuracy. Our resulting models consistently outperform strong baselines across eight benchmark datasets, with an F1 relative improvement of 2.5-130.9% (median 24.4%). More importantly, in contrast to previous reports, our results generalise across external datasets, cohorts, living environments, and sensor devices. Our open-sourced pre-trained models will be valuable in domains with limited labelled data or where good sampling coverage (across devices, populations, and activities) is hard to achieve.
Non-contact physiological monitoring of preterm infants in the Neonatal Intensive Care Unit.
The implementation of video-based non-contact technologies to monitor the vital signs of preterm infants in the hospital presents several challenges, such as the detection of the presence or the absence of a patient in the video frame, robustness to changes in lighting conditions, automated identification of suitable time periods and regions of interest from which vital signs can be estimated. We carried out a clinical study to evaluate the accuracy and the proportion of time that heart rate and respiratory rate can be estimated from preterm infants using only a video camera in a clinical environment, without interfering with regular patient care. A total of 426.6 h of video and reference vital signs were recorded for 90 sessions from 30 preterm infants in the Neonatal Intensive Care Unit (NICU) of the John Radcliffe Hospital in Oxford. Each preterm infant was recorded under regular ambient light during daytime for up to four consecutive days. We developed multi-task deep learning algorithms to automatically segment skin areas and to estimate vital signs only when the infant was present in the field of view of the video camera and no clinical interventions were undertaken. We propose signal quality assessment algorithms for both heart rate and respiratory rate to discriminate between clinically acceptable and noisy signals. The mean absolute error between the reference and camera-derived heart rates was 2.3 beats/min for over 76% of the time for which the reference and camera data were valid. The mean absolute error between the reference and camera-derived respiratory rate was 3.5 breaths/min for over 82% of the time. Accurate estimates of heart rate and respiratory rate could be derived for at least 90% of the time, if gaps of up to 30 seconds with no estimates were allowed.
Nanopore sequencing of influenza A and B in Oxfordshire and the United Kingdom, 2022-23.
OBJECTIVES: We evaluated Nanopore sequencing for influenza surveillance. METHODS: Influenza A and B PCR-positive samples from hospital patients in Oxfordshire, UK, and a UK-wide population survey from winter 2022-23 underwent Nanopore sequencing following targeted rt-PCR amplification. RESULTS: From 941 infections, successful sequencing was achieved in 292/388(75%) available Oxfordshire samples: 231(79%) A/H3N2, 53(18%) A/H1N1, and 8(3%) B/Victoria and in 53/113(47%) UK-wide samples. Sequencing was more successful at lower Ct values. Most same-sample replicate sequences had identical haemagglutinin segments (124/141;88%); 36/39(92%) Illumina vs. Nanopore comparisons were identical, and 3(8%) differed by 1 variant. Comparison of Oxfordshire and UK-wide sequences showed frequent inter-regional transmission. Infections were closely-related to 2022-23 vaccine strains. Only one sample had a neuraminidase inhibitor resistance mutation. 849/941(90%) Oxfordshire infections were community-acquired. 63/88(72%) potentially healthcare-associated cases shared a hospital ward with ≥1 known infectious case. 33 epidemiologically-plausible transmission links had sequencing data for both source and recipient: 8 were within ≤5 SNPs, of these, 5(63%) involved potential sources that were also hospital-acquired. CONCLUSIONS: Nanopore influenza sequencing was reproducible and antiviral resistance rare. Inter-regional transmission was common; most infections were genomically similar. Hospital-acquired infections are likely an important source of nosocomial transmission and should be prioritised for infection prevention and control.
Genome-wide analysis in over 1 million individuals of European ancestry yields improved polygenic risk scores for blood pressure traits.
Hypertension affects more than one billion people worldwide. Here we identify 113 novel loci, reporting a total of 2,103 independent genetic signals (P -8) from the largest single-stage blood pressure (BP) genome-wide association study to date (n = 1,028,980 European individuals). These associations explain more than 60% of single nucleotide polymorphism-based BP heritability. Comparing top versus bottom deciles of polygenic risk scores (PRSs) reveals clinically meaningful differences in BP (16.9 mmHg systolic BP, 95% CI, 15.5-18.2 mmHg, P = 2.22 × 10-126) and more than a sevenfold higher odds of hypertension risk (odds ratio, 7.33; 95% CI, 5.54-9.70; P = 4.13 × 10-44) in an independent dataset. Adding PRS into hypertension-prediction models increased the area under the receiver operating characteristic curve (AUROC) from 0.791 (95% CI, 0.781-0.801) to 0.826 (95% CI, 0.817-0.836, ∆AUROC, 0.035, P = 1.98 × 10-34). We compare the 2,103 loci results in non-European ancestries and show significant PRS associations in a large African-American sample. Secondary analyses implicate 500 genes previously unreported for BP. Our study highlights the role of increasingly large genomic studies for precision health research.
Leveraging functional genomic annotations and genome coverage to improve polygenic prediction of complex traits within and between ancestries.
We develop a method, SBayesRC, that integrates genome-wide association study (GWAS) summary statistics with functional genomic annotations to improve polygenic prediction of complex traits. Our method is scalable to whole-genome variant analysis and refines signals from functional annotations by allowing them to affect both causal variant probability and causal effect distribution. We analyze 50 complex traits and diseases using ∼7 million common single-nucleotide polymorphisms (SNPs) and 96 annotations. SBayesRC improves prediction accuracy by 14% in European ancestry and up to 34% in cross-ancestry prediction compared to the baseline method SBayesR, which does not use annotations, and outperforms other methods, including LDpred2, LDpred-funct, MegaPRS, PolyPred-S and PRS-CSx. Investigation of factors affecting prediction accuracy identifies a significant interaction between SNP density and annotation information, suggesting whole-genome sequence variants with annotations may further improve prediction. Functional partitioning analysis highlights a major contribution of evolutionary constrained regions to prediction accuracy and the largest per-SNP contribution from nonsynonymous SNPs.
Phenomewide Association Study of Health Outcomes Associated With the Genetic Correlates of 25 Hydroxyvitamin D Concentration and Vitamin D Binding Protein Concentration.
While it is known that vitamin D deficiency is associated with adverse bone outcomes, it remains unclear whether low vitamin D status may increase the risk of a wider range of health outcomes. We had the opportunity to explore the association between common genetic variants associated with both 25 hydroxyvitamin D (25OHD) and the vitamin D binding protein (DBP, encoded by the GC gene) with a comprehensive range of health disorders and laboratory tests in a large academic medical center. We used summary statistics for 25OHD and DBP to generate polygenic scores (PGS) for 66,482 participants with primarily European ancestry and 13,285 participants with primarily African ancestry from the Vanderbilt University Medical Center Biobank (BioVU). We examined the predictive properties of PGS25OHD, and two scores related to DBP concentration with respect to 1322 health-related phenotypes and 315 laboratory-measured phenotypes from electronic health records. In those with European ancestry: (a) the PGS25OHD and PGSDBP scores, and individual SNPs rs4588 and rs7041 were associated with both 25OHD concentration and 1,25 dihydroxyvitamin D concentrations; (b) higher PGS25OHD was associated with decreased concentrations of triglycerides and cholesterol, and reduced risks of vitamin D deficiency, disorders of lipid metabolism, and diabetes. In general, the findings for the African ancestry group were consistent with findings from the European ancestry analyses. Our study confirms the utility of PGS and two key variants within the GC gene (rs4588 and rs7041) to predict the risk of vitamin D deficiency in clinical settings and highlights the shared biology between vitamin D-related genetic pathways a range of health outcomes.
[Guidelines for clinical trial protocols for interventions involving artificial intelligence: the SPIRIT-AI extensionDiretrizes para protocolos de ensaios clínicos com intervenções que utilizam inteligência artificial: a extensão SPIRIT-AI].
The SPIRIT 2013 statement aims to improve the completeness of clinical trial protocol reporting by providing evidence-based recommendations for the minimum set of items to be addressed. This guidance has been instrumental in promoting transparent evaluation of new interventions. More recently, there has been a growing recognition that interventions involving artificial intelligence (AI) need to undergo rigorous, prospective evaluation to demonstrate their impact on health outcomes. The SPIRIT-AI (Standard Protocol Items: Recommendations for Interventional Trials-Artificial Intelligence) extension is a new reporting guideline for clinical trial protocols evaluating interventions with an AI component. It was developed in parallel with its companion statement for trial reports: CONSORT-AI (Consolidated Standards of Reporting Trials-Artificial Intelligence). Both guidelines were developed through a staged consensus process involving literature review and expert consultation to generate 26 candidate items, which were consulted upon by an international multi-stakeholder group in a two-stage Delphi survey (103 stakeholders), agreed upon in a consensus meeting (31 stakeholders) and refined through a checklist pilot (34 participants). The SPIRIT-AI extension includes 15 new items that were considered sufficiently important for clinical trial protocols of AI interventions. These new items should be routinely reported in addition to the core SPIRIT 2013 items. SPIRIT-AI recommends that investigators provide clear descriptions of the AI intervention, including instructions and skills required for use, the setting in which the AI intervention will be integrated, considerations for the handling of input and output data, the human-AI interaction and analysis of error cases. SPIRIT-AI will help promote transparency and completeness for clinical trial protocols for AI interventions. Its use will assist editors and peer reviewers, as well as the general readership, to understand, interpret and critically appraise the design and risk of bias for a planned clinical trial.
[Reporting guidelines for clinical trial reports for interventions involving artificial intelligence: the CONSORT-AI extensionDiretrizes para relatórios de ensaios clínicos com intervenções que utilizam inteligência artificial: a extensão CONSORT-AI].
The CONSORT 2010 statement provides minimum guidelines for reporting randomized trials. Its widespread use has been instrumental in ensuring transparency in the evaluation of new interventions. More recently, there has been a growing recognition that interventions involving artificial intelligence (AI) need to undergo rigorous, prospective evaluation to demonstrate impact on health outcomes. The CONSORT-AI (Consolidated Standards of Reporting Trials-Artificial Intelligence) extension is a new reporting guideline for clinical trials evaluating interventions with an AI component. It was developed in parallel with its companion statement for clinical trial protocols: SPIRIT-AI (Standard Protocol Items: Recommendations for Interventional Trials-Artificial Intelligence). Both guidelines were developed through a staged consensus process involving literature review and expert consultation to generate 29 candidate items, which were assessed by an international multi-stakeholder group in a two-stage Delphi survey (103 stakeholders), agreed upon in a two-day consensus meeting (31 stakeholders) and refined through a checklist pilot (34 participants). The CONSORT-AI extension includes 14 new items that were considered sufficiently important for AI interventions that they should be routinely reported in addition to the core CONSORT 2010 items. CONSORT-AI recommends that investigators provide clear descriptions of the AI intervention, including instructions and skills required for use, the setting in which the AI intervention is integrated, the handling of inputs and outputs of the AI intervention, the human-AI interaction and provision of an analysis of error cases. CONSORT-AI will help promote transparency and completeness in reporting clinical trials for AI interventions. It will assist editors and peer reviewers, as well as the general readership, to understand, interpret and critically appraise the quality of clinical trial design and risk of bias in the reported outcomes.
Differentially Private Statistical Inference through β-Divergence One Posterior Sampling
Differential privacy guarantees allow the results of a statistical analysis involving sensitive data to be released without compromising the privacy of any individual taking part. Achieving such guarantees generally requires the injection of noise, either directly into parameter estimates or into the estimation process. Instead of artificially introducing perturbations, sampling from Bayesian posterior distributions has been shown to be a special case of the exponential mechanism, producing consistent, and efficient private estimates without altering the data generative process. The application of current approaches has, however, been limited by their strong bounding assumptions which do not hold for basic models, such as simple linear regressors. To ameliorate this, we propose βD-Bayes, a posterior sampling scheme from a generalised posterior targeting the minimisation of the β-divergence between the model and the data generating process. This provides private estimation that is generally applicable without requiring changes to the underlying model and consistently learns the data generating parameter. We show that βD-Bayes produces more precise inference estimation for the same privacy guarantees, and further facilitates differentially private estimation via posterior sampling for complex classifiers and continuous regression models such as neural networks for the first time.
Differences in 5'untranslated regions highlight the importance of translational regulation of dosage sensitive genes.
Untranslated regions (UTRs) are important mediators of post-transcriptional regulation. The length of UTRs and the composition of regulatory elements within them are known to vary substantially across genes, but little is known about the reasons for this variation in humans. Here, we set out to determine whether this variation, specifically in 5'UTRs, correlates with gene dosage sensitivity. We investigate 5'UTR length, the number of alternative transcription start sites, the potential for alternative splicing, the number and type of upstream open reading frames (uORFs) and the propensity of 5'UTRs to form secondary structures. We explore how these elements vary by gene tolerance to loss-of-function (LoF; using the LOEUF metric), and in genes where changes in dosage are known to cause disease. We show that LOEUF correlates with 5'UTR length and complexity. Genes that are most intolerant to LoF have longer 5'UTRs, greater TSS diversity, and more upstream regulatory elements than their LoF tolerant counterparts. We show that these differences are evident in disease gene-sets, but not in recessive developmental disorder genes where LoF of a single allele is tolerated. Our results confirm the importance of post-transcriptional regulation through 5'UTRs in tight regulation of mRNA and protein levels, particularly for genes where changes in dosage are deleterious and lead to disease. Finally, to support gene-based investigation we release a web-based browser tool, VuTR, that supports exploration of the composition of individual 5'UTRs and the impact of genetic variation within them.
CamTrapAsia: A dataset of tropical forest vertebrate communities from 239 camera trapping studies.
Information on tropical Asian vertebrates has traditionally been sparse, particularly when it comes to cryptic species inhabiting the dense forests of the region. Vertebrate populations are declining globally due to land-use change and hunting, the latter frequently referred as "defaunation." This is especially true in tropical Asia where there is extensive land-use change and high human densities. Robust monitoring requires that large volumes of vertebrate population data be made available for use by the scientific and applied communities. Camera traps have emerged as an effective, non-invasive, widespread, and common approach to surveying vertebrates in their natural habitats. However, camera-derived datasets remain scattered across a wide array of sources, including published scientific literature, gray literature, and unpublished works, making it challenging for researchers to harness the full potential of cameras for ecology, conservation, and management. In response, we collated and standardized observations from 239 camera trap studies conducted in tropical Asia. There were 278,260 independent records of 371 distinct species, comprising 232 mammals, 132 birds, and seven reptiles. The total trapping effort accumulated in this data paper consisted of 876,606 trap nights, distributed among Indonesia, Singapore, Malaysia, Bhutan, Thailand, Myanmar, Cambodia, Laos, Vietnam, Nepal, and far eastern India. The relatively standardized deployment methods in the region provide a consistent, reliable, and rich count data set relative to other large-scale pressence-only data sets, such as the Global Biodiversity Information Facility (GBIF) or citizen science repositories (e.g., iNaturalist), and is thus most similar to eBird. To facilitate the use of these data, we also provide mammalian species trait information and 13 environmental covariates calculated at three spatial scales around the camera survey centroids (within 10-, 20-, and 30-km buffers). We will update the dataset to include broader coverage of temperate Asia and add newer surveys and covariates as they become available. This dataset unlocks immense opportunities for single-species ecological or conservation studies as well as applied ecology, community ecology, and macroecology investigations. The data are fully available to the public for utilization and research. Please cite this data paper when utilizing the data.
Recommendations for laboratory workflow that better support centralised amalgamation of genomic variant data: findings from CanVIG-UK national molecular laboratory survey.
BackgroundNational and international amalgamation of genomic data offers opportunity for research and audit, including analyses enabling improved classification of variants of uncertain significance. Review of individual-level data from National Health Service (NHS) testing of cancer susceptibility genes (2002-2023) submitted to the National Disease Registration Service revealed heterogeneity across participating laboratories regarding (1) the structure, quality and completeness of submitted data, and (2) the ease with which that data could be assembled locally for submission.MethodsIn May 2023, we undertook a closed online survey of 51 clinical scientists who provided consensus responses representing all 17 of 17 NHS molecular genetic laboratories in England and Wales which undertake NHS diagnostic analyses of cancer susceptibility genes. The survey included 18 questions relating to 'next-generation sequencing workflow' (11), 'variant classification' (3) and 'phenotypical context' (4).ResultsWidely differing processes were reported for transfer of variant data into their local LIMS (Laboratory Information Management System), for the formatting in which the variants are stored in the LIMS and which classes of variants are retained in the local LIMS. Differing local provisions and workflow for variant classifications were also reported, including the resources provided and the mechanisms by which classifications are stored.ConclusionThe survey responses illustrate heterogeneous laboratory workflow for preparation of genomic variant data from local LIMS for centralised submission. Workflow is often labour-intensive and inefficient, involving multiple manual steps which introduce opportunities for error. These survey findings and adoption of the concomitant recommendations may support improvement in laboratory dataflows, better facilitating submission of data for central amalgamation.
Image-based consensus molecular subtyping in rectal cancer biopsies and response to neoadjuvant chemoradiotherapy.
The development of deep learning (DL) models to predict the consensus molecular subtypes (CMS) from histopathology images (imCMS) is a promising and cost-effective strategy to support patient stratification. Here, we investigate whether imCMS calls generated from whole slide histopathology images (WSIs) of rectal cancer (RC) pre-treatment biopsies are associated with pathological complete response (pCR) to neoadjuvant long course chemoradiotherapy (LCRT) with single agent fluoropyrimidine. DL models were trained to classify WSIs of colorectal cancers stained with hematoxylin and eosin into one of the four CMS classes using a multi-centric dataset of resection and biopsy specimens (n = 1057 WSIs) with paired transcriptional data. Classifiers were tested on a held out RC biopsy cohort (ARISTOTLE) and correlated with pCR to LCRT in an independent dataset merging two RC cohorts (ARISTOTLE, n = 114 and SALZBURG, n = 55 patients). DL models predicted CMS with high classification performance in multiple comparative analyses. In the independent cohorts (ARISTOTLE, SALZBURG), cases with WSIs classified as imCMS1 had a significantly higher likelihood of achieving pCR (OR = 2.69, 95% CI 1.01-7.17, p = 0.048). Conversely, imCMS4 was associated with lack of pCR (OR = 0.25, 95% CI 0.07-0.88, p = 0.031). Classification maps demonstrated pathologist-interpretable associations with high stromal content in imCMS4 cases, associated with poor outcome. No significant association was found in imCMS2 or imCMS3. imCMS classification of pre-treatment biopsies is a fast and inexpensive solution to identify patient groups that could benefit from neoadjuvant LCRT. The significant associations between imCMS1/imCMS4 with pCR suggest the existence of predictive morphological features that could enhance standard pathological assessment.