Using a patient-reported outcome measure to assess quality of life at Western Sydney Sexual Health Centre.

Background Attaining a good quality of life (QOL) is a priority for people living with HIV (PLHIV). We explored the interaction between QOL and the associated demographic, behavioural and clinical factors for PLHIV attending an outer-metropolitan clinical setting in Western Sydney, Australia. The clinic's cohort of PLHIV is characterised by relatively high proportions of women, heterosexual men, and patients from culturally and linguistically diverse (CALD) communities. Methods We assessed QOL using the PozQol tool that is specifically designed for PLHIV. QOL scores and de-identified socio-demographic and clinical data were extracted from the electronic and paper medical records of PLHIV who completed a PozQol tool (September 2020-March 2022). We performed descriptive analyses and logistic regression to identify associations. Results Among 188 patients, there were 77.7% men, 21.3% women, 1.1% transwomen; 67.0% were born overseas, 85.1% spoke English, 84.4% were Medicare-eligible, 85.9% were employed, 58.5% were diagnosed with HIV 6-20years ago, and 33.0% within the past 5years. Overall, 58.0% had a high or very high QOL. A low score in any domain was associated with Medicare-ineligibility. Low QOL scores in specific domains were associated with the following factors: health (being born overseas, having partners of both sexes), psychological (unemployment, having a mental health condition, having a viral load >20 copies/mL), social (unemployment), and functional (Medicare-eligibility, unemployment, having a viral load >20 copies/mL). Conclusions The PozQol tool has enhanced understanding of factors impacting on QOL for PLHIV attending our service in Western Sydney. Identifying patients with low QOL scores allows targeted clinical interventions to improve QOL, and re-alignment of clinical services to better support PLHIV.

Discovery of PLD4 modulators by high-throughput screening and kinetic analysis.

Phospholipase D3 (PLD3) and D4 (PLD4) are endolysosomal exonucleases of ssDNA and ssRNA that regulate innate immunity. Polymorphisms of these enzymes are correlated with numerous human diseases, including Alzheimer's, rheumatoid arthritis, and systemic sclerosis. Pharmacological modulation of these immunoregulatory proteins may yield novel immunotherapies and adjuvants. A previous study reported a high-throughput screen (N = 17,952) that discovered a PLD3-selective activator and inhibitor, as well as a nonselective inhibitor, but failed to identify selective modulators of PLD4. However, modulators selective for PLD4 are therapeutically pertinent, since recent reports have shown that regulating this protein has direct implications in cancer and autoimmune diseases. Furthermore, the high expression of PLD4 in dendritic and myeloid cells, in comparison to the broader expression of PLD3, presents the opportunity for a cell-targeted immunotherapy. Here, we describe screening of an expended diversity library (N = 45,760) with an improved platform and report the discovery of one inhibitor and three activators selective for PLD4. Furthermore, kinetic modeling and structural analysis suggest mechanistic differences in the modulation of these hits. These findings further establish the utility of this screening platform and provide a set of chemical scaffolds to guide future small-molecule development for this novel immunoregulator target.

Comparison of exposure of the radius afforded by the craniomedial and craniolateral open surgical approaches in dog cadavers.

Management of antebrachial fractures in dogs typically involves open reduction and bone plate stabilisation of the radius performed using either a craniomedial approach (CMA) or craniolateral approach (CLA). The aim of this cadaveric study was to compare radial exposure afforded via the craniomedial and craniolateral approaches, without and with releasing the abductor digiti I longus (ADIL) muscle. The CMA and CLA were performed on alternate paired cadaver forelimbs with the subsequent release (CMA) or elevation (CLA) of the ADIL muscle (n = 12). The percent area of exposure was calculated in defined longitudinal radial segments (LRSs) and hemi-circumferential radial segments (HRSs) using digitally uploaded images. A non-parametric rank sum test was used for statistical analysis (p < 0.05). While both the CMA and CLA provided sufficient exposure for radial cranial plate placement in the mid-to-distal diaphyseal region (50-80% LRSs), the CMA afforded greater exposure to the distal metaphysis (80-90% LRS). The CLA was advantageous in exposing the proximal diaphysis (20-50% LRSs) and distal epiphysis (90-100% LRS). Exposure of the distal diaphyseal, metaphyseal, and epiphyseal regions (80-100% LRSs) was impeded by the ADIL muscle. A tenotomy of the tendon of insertion of the ADIL muscle increased exposure afforded by the CMA in the distal metaphyseal and epiphyseal regions. The exposure afforded by the CLA in the distal diaphyseal and metaphyseal regions was significantly improved by elevating the origin of the ADIL muscle. The results of this study provide useful clinical information when surgeons are deciding which approach to use for cranial radial plating.

Allopurinol and cardiovascular outcomes in patients with ischaemic heart disease: the ALL-HEART RCT and economic evaluation.

Background: Allopurinol is a xanthine oxidase inhibitor that lowers serum uric acid and is used to prevent acute gout flares in patients with gout. Observational and small interventional studies have suggested beneficial cardiovascular effects of allopurinol. Objective: To determine whether allopurinol improves major cardiovascular outcomes in patients with ischaemic heart disease. Design: Prospective, randomised, open-label, blinded endpoint multicentre clinical trial. Setting: Four hundred and twenty-four UK primary care practices. Participants: Aged 60 years and over with ischaemic heart disease but no gout. Interventions: Participants were randomised (1 : 1) using a central web-based randomisation system to receive allopurinol up to 600 mg daily that was added to usual care or to continue usual care. Main outcome measures: The primary outcome was the composite of non-fatal myocardial infarction, non-fatal stroke or cardiovascular death. Secondary outcomes were non-fatal myocardial infarction, non-fatal stroke, cardiovascular death, all-cause mortality, hospitalisation for heart failure, hospitalisation for acute coronary syndrome, coronary revascularisation, hospitalisation for acute coronary syndrome or coronary revascularisation, all cardiovascular hospitalisations, quality of life and cost-effectiveness. The hazard ratio (allopurinol vs. usual care) in a Cox proportional hazards model was assessed for superiority in a modified intention-to-treat analysis. Results: From 7 February 2014 to 2 October 2017, 5937 participants were enrolled and randomised to the allopurinol arm (n = 2979) or the usual care arm (n = 2958). A total of 5721 randomised participants (2853 allopurinol; 2868 usual care) were included in the modified intention-to-treat analysis population (mean age 72.0 years; 75.5% male). There was no difference between the allopurinol and usual care arms in the primary endpoint, 314 (11.0%) participants in the allopurinol arm (2.47 events per 100 patient-years) and 325 (11.3%) in the usual care arm (2.37 events per 100 patient-years), hazard ratio 1.04 (95% confidence interval 0.89 to 1.21); p = 0.65. Two hundred and eighty-eight (10.1%) participants in the allopurinol arm and 303 (10.6%) participants in the usual care arm died, hazard ratio 1.02 (95% confidence interval 0.87 to 1.20); p = 0.77. The pre-specified health economic analysis plan was to perform a 'within trial' cost-utility analysis if there was no statistically significant difference in the primary endpoint, so NHS costs and quality-adjusted life-years were estimated over a 5-year period. The difference in costs between treatment arms was +£115 higher for allopurinol (95% confidence interval £17 to £210) with no difference in quality-adjusted life-years (95% confidence interval -0.061 to +0.060). We conclude that there is no evidence that allopurinol used in line with the study protocol is cost-effective. Limitations: The results may not be generalisable to younger populations, other ethnic groups or patients with more acute ischaemic heart disease. One thousand six hundred and thirty-seven participants (57.4%) in the allopurinol arm withdrew from randomised treatment, but an on-treatment analysis gave similar results to the main analysis. Conclusions: The ALL-HEART study showed that treatment with allopurinol 600 mg daily did not improve cardiovascular outcomes compared to usual care in patients with ischaemic heart disease. We conclude that allopurinol should not be recommended for the secondary prevention of cardiovascular events in patients with ischaemic heart disease but no gout. Future work: The effects of allopurinol on cardiovascular outcomes in patients with ischaemic heart disease and co-existing hyperuricaemia or clinical gout could be explored in future studies. Trial registration: This trial is registered as EU Clinical Trials Register (EudraCT 2013-003559-39) and ISRCTN (ISRCTN 32017426). Funding: This award was funded by the National Institute for Health and Care Research (NIHR) Health Technology Assessment programme (NIHR award ref: 11/36/41) and is published in full in Health Technology Assessment; Vol. 28, No. 18. See the NIHR Funding and Awards website for further award information.

The purpose of the ALL-HEART study was to determine whether giving allopurinol to people with ischaemic heart disease (also commonly known as coronary heart disease) would reduce their risk of having a heart attack, stroke or of dying from cardiovascular disease. Allopurinol is a medication usually given to patients with gout to prevent acute gout flares. It is not currently used to treat ischaemic heart disease. We randomly allocated people aged over 60 years with ischaemic heart disease to take up to 600 mg of allopurinol daily (in addition to their usual care) or to continue with their usual care. We then monitored participants for several years and recorded any major health events such as heart attacks, strokes and deaths. We obtained most of the follow-up data from centrally held electronic hospital admissions and death records, making the study easier for participants and more cost-efficient. We asked participants in both groups to complete questionnaires to assess their quality of life during the study. We also collected data to determine whether there was any economic benefit to the NHS of using allopurinol in patients with ischaemic heart disease. There was no difference in the risk of heart attacks, strokes or death from cardiovascular disease between the participants given allopurinol and those in the group continuing their usual care. We also found no difference in the risks of other cardiovascular events, deaths from any cause or quality-of-life measurements between the allopurinol and usual care groups. The results of the ALL-HEART study suggest that we should not recommend that allopurinol be given to people with ischaemic heart disease to prevent further cardiovascular events or deaths.

Continuous ultraviolet to blue-green astrocomb.

Cosmological and exoplanetary science using transformative telescopes like the ELT will demand precise calibration of astrophysical spectrographs in the blue-green, where stellar absorption lines are most abundant. Astrocombs-lasers providing a broadband sequence of regularly-spaced optical frequencies on a multi-GHz grid-promise an atomically-traceable calibration scale, but their realization in the blue-green is challenging for current infrared-laser-based technology. Here, we introduce a concept achieving a broad, continuous spectrum by combining second-harmonic generation and sum-frequency-mixing in an MgO:PPLN waveguide to generate 390-520 nm light from a 1 GHz Ti:sapphire frequency comb. Using a Fabry-Pérot filter, we extract a 30 GHz sub-comb spanning 392-472 nm, visualizing its thousands of modes on a high-resolution spectrograph. Experimental data and simulations demonstrate how the approach can bridge the spectral gap present in second-harmonic-only conversion. Requiring only [Formula: see text]100 pJ pulses, our concept establishes a new route to broadband UV-visible generation at GHz repetition rates.

Total Synthesis as Training for Medicinal Chemistry.

There is an ongoing debate about the best types of training in academia for practicing modern medicinal chemistry in the pharmaceutical and biotechnology industries of today. A case is made in this Viewpoint for the ongoing, and perhaps increasing, value of total synthesis as training for medicinal chemistry.

Pneumocystis jirovecii pneumonia presenting with severe, refractory hypercalcaemia in an immunosuppressed renal transplant patient: Review of the literature and novel biochemical insights.

Pneumocystis jirovecii pneumonia (PJP) is a rare but serious complication of immunosuppression post-solid organ transplantation. We present a case of refractory, severe hypercalcaemia due to PJP in a renal transplant recipient. Treatment of PJP led to normalisation of the patient's calcium levels, and clinical improvement. To further explore the proposed calcitriol-driven mechanism leading to hypercalcaemia in PJP, we performed biochemical analysis on pre- and post-treatment serum and bronchoalveolar lavage sample at the time of PJP diagnosis. We confirmed high circulating and pulmonary levels of calcitriol in acute, untreated PJP with severe hypercalcaemia. PJP treatment led to reduction of circulating calcitriol to within normal range. We present this case, together with a literature review of similar reported cases, and the novel biochemical evidence supporting extra-renal production of calcitriol by activated pulmonary macrophages as the mechanism underpinning hypercalcaemia in PJP.

Defining quality assessment in vascular surgery training: an expert Delphi process.

INTRODUCTION: A robust and reproducible way of assessing training should optimise and standardise vascular surgical training. This study describes the methodology supporting the Vascular Surgery Specialty Advisory Committee regional quality assurance reports for vascular surgery training programmes in the UK. METHODS: A Delphi consensus exercise was performed to establish the domains of training that most appropriately assess the quality of a vascular surgery training programme. A total of 54 stakeholders were invited to participate, including trainees, training programme directors and members of the vascular speciality advisory committee (SAC), vascular society executive and education committees. RESULTS: A total of 39 stakeholders successfully completed the three-stage Delphi process over 15 weeks. The domains identified as most appropriate to assess the quality of a vascular training programme were: Joint Committee on Surgical Training (JCST) survey results, clinical experience, regional education programmes, radiology support, timetable, regional support for trainees, trainer support for trainees, opportunities for professional development, trainee-rated quality of consultant teaching and training, and trainee recommendation of the post. CONCLUSIONS: This study describes a method to identify and prioritise domains that are appropriate to assess the quality of a vascular training programme. The domains that were identified as appropriate to assess quality are transferable internationally and the Delphi methodology could be used by other training schemes to 'fine-tune' their own domains to review and optimise the quality of their own training programmes.

Persistent mucus plugs in proximal airways are consequential for airflow limitation in asthma.

BACKGROUNDInformation about the size, airway location, and longitudinal behavior of mucus plugs in asthma is needed to understand their role in mechanisms of airflow obstruction and to rationally design muco-active treatments.METHODSCT lung scans from 57 patients with asthma were analyzed to quantify mucus plug size and airway location, and paired CT scans obtained 3 years apart were analyzed to determine plug behavior over time. Radiologist annotations of mucus plugs were incorporated in an image-processing pipeline to generate size and location information that was related to measures of airflow.RESULTSThe length distribution of 778 annotated mucus plugs was multimodal, and a 12 mm length defined short ("stubby", ≤12 mm) and long ("stringy", >12 mm) plug phenotypes. High mucus plug burden was disproportionately attributable to stringy mucus plugs. Mucus plugs localized predominantly to airway generations 6-9, and 47% of plugs in baseline scans persisted in the same airway for 3 years and fluctuated in length and volume. Mucus plugs in larger proximal generations had greater effects on spirometry measures than plugs in smaller distal generations, and a model of airflow that estimates the increased airway resistance attributable to plugs predicted a greater effect for proximal generations and more numerous mucus plugs.CONCLUSIONPersistent mucus plugs in proximal airway generations occur in asthma and demonstrate a stochastic process of formation and resolution over time. Proximal airway mucus plugs are consequential for airflow and are in locations amenable to treatment by inhaled muco-active drugs or bronchoscopy.TRIAL REGISTRATIONClinicaltrials.gov; NCT01718197, NCT01606826, NCT01750411, NCT01761058, NCT01761630, NCT01716494, and NCT01760915.FUNDINGAstraZeneca, Boehringer-Ingelheim, Genentech, GlaxoSmithKline, Sanofi-Genzyme-Regeneron, and TEVA provided financial support for study activities at the Coordinating and Clinical Centers beyond the third year of patient follow-up. These companies had no role in study design or data analysis, and the only restriction on the funds was that they be used to support the SARP initiative.

The Alien Flora of Australia (AFA), a unified Australian national dataset on plant invasion.

Biological invasions are a major threat to Australia. Information on alien flora in Australia is collated independently by different jurisdictions, which has led to inconsistencies at the national level, hampering efficient management. To harmonise different information sources, we present the Alien Flora of Australia (AFA), a nationally unified dataset. To create the AFA, we developed an R script that compares existing data sources (the Australian Plant Census and state and territory censuses), identifies mismatches among them and integrates the information into unified invasion statuses at the national scale. The AFA follows the taxonomy and nomenclature adopted for the Australian Plant Census, introduction status and impact of plants known to occur in Australia. The up-to-date information presented in this dataset can aid early warning of alien species invasions, facilitate decision-making at different levels, and biosecurity at national scale. The associated script is ready to be implemented into new versions of the AFA with updated releases of any of the data sources, streamlining future efforts to track of alien flora across Australia.

Warning: Humans cannot reliably detect speech deepfakes.

Speech deepfakes are artificial voices generated by machine learning models. Previous literature has highlighted deepfakes as one of the biggest security threats arising from progress in artificial intelligence due to their potential for misuse. However, studies investigating human detection capabilities are limited. We presented genuine and deepfake audio to n = 529 individuals and asked them to identify the deepfakes. We ran our experiments in English and Mandarin to understand if language affects detection performance and decision-making rationale. We found that detection capability is unreliable. Listeners only correctly spotted the deepfakes 73% of the time, and there was no difference in detectability between the two languages. Increasing listener awareness by providing examples of speech deepfakes only improves results slightly. As speech synthesis algorithms improve and become more realistic, we can expect the detection task to become harder. The difficulty of detecting speech deepfakes confirms their potential for misuse and signals that defenses against this threat are needed.

Deep Learning Synthetic Strain: Quantitative Assessment of Regional Myocardial Wall Motion at MRI.

Purpose: To assess the feasibility of a newly developed algorithm, called deep learning synthetic strain (DLSS), to infer myocardial velocity from cine steady-state free precession (SSFP) images and detect wall motion abnormalities in patients with ischemic heart disease. Materials and Methods: In this retrospective study, DLSS was developed by using a data set of 223 cardiac MRI examinations including cine SSFP images and four-dimensional flow velocity data (November 2017 to May 2021). To establish normal ranges, segmental strain was measured in 40 individuals (mean age, 41 years ± 17 [SD]; 30 men) without cardiac disease. Then, DLSS performance in the detection of wall motion abnormalities was assessed in a separate group of patients with coronary artery disease, and these findings were compared with consensus results of four independent cardiothoracic radiologists (ground truth). Algorithm performance was evaluated by using receiver operating characteristic curve analysis. Results: Median peak segmental radial strain in individuals with normal cardiac MRI findings was 38% (IQR: 30%-48%). Among patients with ischemic heart disease (846 segments in 53 patients; mean age, 61 years ± 12; 41 men), the Cohen κ among four cardiothoracic readers for detecting wall motion abnormalities was 0.60-0.78. DLSS achieved an area under the receiver operating characteristic curve of 0.90. Using a fixed 30% threshold for abnormal peak radial strain, the algorithm achieved a sensitivity, specificity, and accuracy of 86%, 85%, and 86%, respectively. Conclusion: The deep learning algorithm had comparable performance with subspecialty radiologists in inferring myocardial velocity from cine SSFP images and identifying myocardial wall motion abnormalities at rest in patients with ischemic heart disease.Keywords: Neural Networks, Cardiac, MR Imaging, Ischemia/Infarction Supplemental material is available for this article. © RSNA, 2023.

Life-long learning in laboratory animal science and ethics for veterinary and para-veterinary professionals in South Africa.

Veterinary and para-veterinary professionals working in the animal research sector are critical to ensure scientific quality and the humane care and use of animals. However, there are few focused education and training opportunities available for these professionals in South Africa. A survey of veterinarians working in animal research, undertaken by the South African Association for Laboratory Animal Science, identified the need for more advanced education and training opportunities beyond the routine Day 1 Skills currently provided for in undergraduate education. These could be broadly categorised into knowledge and skills relating to species-specific husbandry, procedures and clinical approaches, research-related biosecurity and biosafety, and study-specific ethical and animal welfare considerations. A subsequent workshop, attended by 85 veterinary and para-veterinary professionals in the animal research sector, identified 53 life-long learning needs, each with an associated learning outcome, for this professional community. These were grouped into five overarching themes: Personal development (9); Leadership and management skills (12); Education and training skills (5); Welfare, ethics and clinical skills (20); and Regulations and quality-assurance (7). Of the 53 learning outcomes, 14 were knowledge-based, ten were competencies, and 29 both knowledge and competence. These life-long learning opportunities, if available and implemented, will address important needs of veterinary and paraveterinary professionals in the animal research sector in South Africa. This would empower these professionals, assist in improving animal and human wellbeing, support high-quality ethical science, and maintain public confidence in the sector, thus enabling a more satisfactory career environment.

Learning outcomes for the education and training of laboratory animal caretakers in South Africa.

Education and training is essential for laboratory animal caretakers (LACs), but there are no courses available in South Africa. A national workshop was thus held to collaboratively establish the learning outcomes (LOs) for the education and training (E&T) of LACs. Eighty-five stakeholders from 30 institutions took part in small group discussions interspersed with plenary sessions to draw up the consensus LOs. One-hundred-and-twenty LOs were identified, grouped into the following three main themes and 15 topics: 1) Focus on animals (animal care and husbandry, animal ethics, animal welfare, basic biology, environment); 2) Focus on humans (administration, health and safety, lifelong learning, professionalism, psychological wellbeing); and 3) Focus on systems (biosecurity, equipment, jurisprudence, logistics, and quality management). This E&T framework provides a foundation for a career path in the laboratory animal science field. The psychological (i.e. mental and emotional) wellbeing of LACs forms a noteworthy component of the focus on humans, since working with research animals is stressful and coping mechanisms are needed in order to promote compassion satisfaction and prevent compassion fatigue and burnout. Approximately 75% of the LOs are knowledge-based, while 25% are competencies in practical skills. It is recommended that competencies should be assessed by direct observation of practical/procedural skills, where competence in a procedure or practical task is assessed against predetermined criteria. These LOs are published with the intent that they will promote animal and human wellbeing, support ethical science, maintain public confidence, and in so doing, contribute to a just and civilised society.

Interpretable Machine Learning Models for Phase Prediction in Polymerization-Induced Self-Assembly.

While polymerization-induced self-assembly (PISA) has become a preferred synthetic route toward amphiphilic block copolymer self-assemblies, predicting their phase behavior from experimental design is extremely challenging, requiring time and work-intensive creation of empirical phase diagrams whenever self-assemblies of novel monomer pairs are sought for specific applications. To alleviate this burden, we develop here the first framework for a data-driven methodology for the probabilistic modeling of PISA morphologies based on a selection and suitable adaption of statistical machine learning methods. As the complexity of PISA precludes generating large volumes of training data with in silico simulations, we focus on interpretable low variance methods that can be interrogated for conformity with chemical intuition and that promise to work well with only 592 training data points which we curated from the PISA literature. We found that among the evaluated linear models, generalized additive models, and rule and tree ensembles, all but the linear models show a decent interpolation performance with around 0.2 estimated error rate and 1 bit expected cross entropy loss (surprisal) when predicting the mixture of morphologies formed from monomer pairs already encountered in the training data. When considering extrapolation to new monomer combinations, the model performance is weaker but the best model (random forest) still achieves highly nontrivial prediction performance (0.27 error rate, 1.6 bit surprisal), which renders it a good candidate to support the creation of empirical phase diagrams for new monomers and conditions. Indeed, we find in three case studies that, when used to actively learn phase diagrams, the model is able to select a smart set of experiments that lead to satisfactory phase diagrams after observing only relatively few data points (5-16) for the targeted conditions. The data set as well as all model training and evaluation codes are publicly available through the GitHub repository of the last author.

Transformation of a Metal Chelate into a "Catch and Anchor" Inhibitor of Botulinum A Protease.

Targeting the botulinum neurotoxin light chain (LC) metalloprotease using small-molecule metal chelate inhibitors is a promising approach to counter the effects of the lethal toxin. However, to overcome the pitfalls associated with simple reversible metal chelate inhibitors, it is crucial to investigate alternative scaffolds/strategies. In conjunction with Atomwise Inc., in silico and in vitro screenings were conducted, yielding a number of leads, including a novel 9-hydroxy-4H-pyrido [1,2-a]pyrimidin-4-one (PPO) scaffold. From this structure, an additional series of 43 derivatives were synthesized and tested, resulting in a lead candidate with a Ki of 150 nM in a BoNT/A LC enzyme assay and 17 µM in a motor neuron cell-based assay. These data combined with structure-activity relationship (SAR) analysis and docking led to a bifunctional design strategy, which we termed "catch and anchor" for the covalent inhibition of BoNT/A LC. Kinetic evaluation was conducted on structures prepared from this catch and anchor campaign, providing kinact/Ki values, and rationale for inhibition seen. Covalent modification was validated through additional assays, including an FRET endpoint assay, mass spectrometry, and exhaustive enzyme dialysis. The data presented support the PPO scaffold as a novel candidate for targeted covalent inhibition of BoNT/A LC.

Deep Learning Applied to Raman Spectroscopy for the Detection of Microsatellite Instability/MMR Deficient Colorectal Cancer.

Defective DNA mismatch repair is one pathogenic pathway to colorectal cancer. It is characterised by microsatellite instability which provides a molecular biomarker for its detection. Clinical guidelines for universal testing of this biomarker are not met due to resource limitations; thus, there is interest in developing novel methods for its detection. Raman spectroscopy (RS) is an analytical tool able to interrogate the molecular vibrations of a sample to provide a unique biochemical fingerprint. The resulting datasets are complex and high-dimensional, making them an ideal candidate for deep learning, though this may be limited by small sample sizes. This study investigates the potential of using RS to distinguish between normal, microsatellite stable (MSS) and microsatellite unstable (MSI-H) adenocarcinoma in human colorectal samples and whether deep learning provides any benefit to this end over traditional machine learning models. A 1D convolutional neural network (CNN) was developed to discriminate between healthy, MSI-H and MSS in human tissue and compared to a principal component analysis-linear discriminant analysis (PCA-LDA) and a support vector machine (SVM) model. A nested cross-validation strategy was used to train 30 samples, 10 from each group, with a total of 1490 Raman spectra. The CNN achieved a sensitivity and specificity of 83% and 45% compared to PCA-LDA, which achieved a sensitivity and specificity of 82% and 51%, respectively. These are competitive with existing guidelines, despite the low sample size, speaking to the molecular discriminative power of RS combined with deep learning. A number of biochemical antecedents responsible for this discrimination are also explored, with Raman peaks associated with nucleic acids and collagen being implicated.