In Vivo Raman Spectroscopy of Muscle Is Highly Sensitive for Detection of Healthy Muscle and Highly Specific for Detection of Disease.

Raman spectroscopy of muscle provides a molecular fingerprint to identify the disease. Previous work has demonstrated effectiveness in differentiating between two groups of equal sizes (e.g., healthy vs disease) but imbalanced multiclass scenarios are more common in medicine. We performed in vivo Raman spectroscopy in a total of 151 mice across four different histopathologies (healthy, acute myopathy, chronic myopathy, neurogenic), with variable numbers in each (class "imbalance"). Using hierarchical modeling and synthetic data generation, we demonstrate high sensitivity (94%) for detection of healthy muscle and high specificity (≥97%) for disease. Further, we demonstrate the potential for unique biomarker development by demonstrating variations in the protein structure across different pathologies. The findings demonstrate the potential of Raman spectroscopy to provide accurate disease identification and unique molecular insights.

Conformational fingerprinting with Raman spectroscopy reveals protein structure as a translational biomarker of muscle pathology.

Neuromuscular disorders are a group of conditions that can result in weakness of skeletal muscles. Examples include fatal diseases such as amyotrophic lateral sclerosis and conditions associated with high morbidity such as myopathies (muscle diseases). Many of these disorders are known to have abnormal protein folding and protein aggregates. Thus, easy to apply methods for the detection of such changes may prove useful diagnostic biomarkers. Raman spectroscopy has shown early promise in the detection of muscle pathology in neuromuscular disorders and is well suited to characterising the conformational profiles relating to protein secondary structure. In this work, we assess if Raman spectroscopy can detect differences in protein structure in muscle in the setting of neuromuscular disease. We utilise in vivo Raman spectroscopy measurements from preclinical models of amyotrophic lateral sclerosis and the myopathy Duchenne muscular dystrophy, together with ex vivo measurements of human muscle samples from individuals with and without myopathy. Using quantitative conformation profiling and matrix factorisation we demonstrate that quantitative 'conformational fingerprinting' can be used to identify changes in protein folding in muscle. Notably, myopathic conditions in both preclinical models and human samples manifested a significant reduction in α-helix structures, with concomitant increases in ß-sheet and, to a lesser extent, nonregular configurations. Spectral patterns derived through non-negative matrix factorisation were able to identify myopathy with a high accuracy (79% in mouse, 78% in human tissue). This work demonstrates the potential of conformational fingerprinting as an interpretable biomarker for neuromuscular disorders.

Combining electromyography and Raman spectroscopy: optical EMG.

INTRODUCTION/AIMS: Electromyography (EMG) remains a key component of the diagnostic work-up for suspected neuromuscular disease, but it does not provide insight into the molecular composition of muscle which can provide diagnostic information. Raman spectroscopy is an emerging neuromuscular biomarker capable of generating highly specific, molecular fingerprints of tissue. Here, we present "optical EMG," a combination of EMG and Raman spectroscopy, achieved using a single needle. METHODS: An optical EMG needle was created to collect electrophysiological and Raman spectroscopic data during a single insertion. We tested functionality with in vivo recordings in the SOD1G93A mouse model of amyotrophic lateral sclerosis (ALS), using both transgenic (n = 10) and non-transgenic (NTg, n = 7) mice. Under anesthesia, compound muscle action potentials (CMAPs), spontaneous EMG activity and Raman spectra were recorded from both gastrocnemius muscles with the optical EMG needle. Standard concentric EMG needle recordings were also undertaken. Electrophysiological data were analyzed with standard univariate statistics, Raman data with both univariate and multivariate analyses. RESULTS: A significant difference in CMAP amplitude was observed between SOD1G93A and NTg mice with optical EMG and standard concentric needles (p = .015 and p = .011, respectively). Spontaneous EMG activity (positive sharp waves) was detected in transgenic SOD1G93A mice only. Raman spectra demonstrated peaks associated with key muscle components. Significant differences in molecular composition between SOD1G93A and NTg muscle were identified through the Raman spectra. DISCUSSION: Optical EMG can provide standard electrophysiological data and molecular Raman data during a single needle insertion and represents a potential biomarker for neuromuscular disease.

Fiber optic Raman spectroscopy for the evaluation of disease state in Duchenne muscular dystrophy: An assessment using the mdx model and human muscle.

INTRODUCTION/AIMS: Raman spectroscopy is an emerging technique for the evaluation of muscle disease. In this study we evaluate the ability of in vivo intramuscular Raman spectroscopy to detect the effects of voluntary running in the mdx model of Duchenne muscular dystrophy (DMD). We also compare mdx data with muscle spectra from human DMD patients. METHODS: Thirty 90-day-old mdx mice were randomly allocated to an exercised group (48-hour access to a running wheel) and an unexercised group (n = 15 per group). In vivo Raman spectra were collected from both gastrocnemius muscles and histopathological assessment subsequently performed. Raman data were analyzed using principal component analysis-fed linear discriminant analysis (PCA-LDA). Exercised and unexercised mdx muscle spectra were compared with human DMD samples using cosine similarity. RESULTS: Exercised mice ran an average of 6.5 km over 48 hours, which induced a significant increase in muscle necrosis (P = .03). PCA-LDA scores were significantly different between the exercised and unexercised groups (P < .0001) and correlated significantly with distance run (P = .01). Raman spectra from exercised mice more closely resembled human spectra than those from unexercised mice. DISCUSSION: Raman spectroscopy provides a readout of the biochemical alterations in muscle in both the mdx mouse and human DMD muscle.

Reinnervation as measured by the motor unit size index is associated with preservation of muscle strength in amyotrophic lateral sclerosis, but not all muscles reinnervate.

INTRODUCTION/AIMS: The motor unit size index (MUSIX) may provide insight into reinnervation patterns in diseases such as amyotrophic lateral sclerosis (ALS). However, it is not known whether MUSIX detects clinically relevant changes in reinnervation, or if all muscles manifest changes in MUSIX in response to reinnervation after motor unit loss. METHODS: Fifty-seven patients with ALS were assessed at 3-month intervals for 12 months in four centers. Muscles examined were abductor pollicis brevis, abductor digiti minimi, biceps brachii, and tibialis anterior. Results were split into two groups: muscles with increases in MUSIX and those without increases. Longitudinal changes in MUSIX, motor unit number index (MUNIX), compound muscle action potential amplitude, and Medical Research Council strength score were investigated. RESULTS: One hundred thirty-three muscles were examined. Fifty-nine percent of the muscles exhibited an increase in MUSIX during the study. Muscles with MUSIX increases lost more motor units (58% decline in MUNIX at 12 months, P < .001) than muscles that did not increase MUSIX (34.6% decline in MUNIX at 12 months, P < .001). However, longitudinal changes in muscle strength were similar. When motor unit loss was similar, the absence of a MUSIX increase was associated with a significantly greater loss of muscle strength (P = .002). DISCUSSION: MUSIX increases are associated with greater motor unit loss but relative preservation of muscle strength. Thus, MUSIX appears to be measuring a clinically relevant response that can provide a quantitative outcome measure of reinnervation in clinical trials. Furthermore, MUSIX suggests that reinnervation may play a major role in determining the progression of weakness.

Label-free fibre optic Raman spectroscopy with bounded simplex-structured matrix factorization for the serial study of serum in amyotrophic lateral sclerosis.

Amyotrophic lateral sclerosis (ALS) is an incurable neurodegenerative disease in urgent need of disease biomarkers for the assessment of promising therapeutic candidates in clinical trials. Raman spectroscopy is an attractive technique for identifying disease related molecular changes due to its simplicity. Here, we describe a fibre optic fluid cell for undertaking spontaneous Raman spectroscopy studies of human biofluids that is suitable for use away from a standard laboratory setting. Using this system, we examined serum obtained from patients with ALS at their first presentation to our centre (n = 66) and 4 months later (n = 27). We analysed Raman spectra using bounded simplex-structured matrix factorization (BSSMF), a generalisation of non-negative matrix factorisation which uses the distribution of the original data to limit the factorisation modes (spectral patterns). Biomarkers associated with ALS disease such as measures of symptom severity, respiratory function and inflammatory/immune pathways (C3/C-reactive protein) correlated with baseline Raman modes. Between visit spectral changes were highly significant (p = 0.0002) and were related to protein structure. Comparison of Raman data with established ALS biomarkers as a trial outcome measure demonstrated a reduction in required sample size with BSSMF Raman. Our portable, simple to use fibre optic system allied to BSSMF shows promise in the quantification of disease-related changes in ALS over short timescales.

Rapid identification of human muscle disease with fibre optic Raman spectroscopy.

The diagnosis of muscle disorders ("myopathies") can be challenging and new biomarkers of disease are required to enhance clinical practice and research. Despite advances in areas such as imaging and genomic medicine, muscle biopsy remains an important but time-consuming investigation. Raman spectroscopy is a vibrational spectroscopy application that could provide a rapid analysis of muscle tissue, as it requires no sample preparation and is simple to perform. Here, we investigated the feasibility of using a miniaturised, portable fibre optic Raman system for the rapid identification of muscle disease. Samples were assessed from 27 patients with a final clinico-pathological diagnosis of a myopathy and 17 patients in whom investigations and clinical follow-up excluded myopathy. Multivariate classification techniques achieved accuracies ranging between 71-77%. To explore the potential of Raman spectroscopy to identify different myopathies, patients were subdivided into mitochondrial and non-mitochondrial myopathy groups. Classification accuracies were between 74-89%. Observed spectral changes were related to changes in protein structure. These data indicate fibre optic Raman spectroscopy is a promising technique for the rapid identification of muscle disease that could provide real time diagnostic information. The application of fibre optic Raman technology raises the prospect of in vivo bedside testing for muscle diseases which would significantly streamline the diagnostic pathway of these disorders.

Magnetic resonance spectroscopy reveals mitochondrial dysfunction in amyotrophic lateral sclerosis.

Mitochondrial dysfunction is postulated to be central to amyotrophic lateral sclerosis (ALS) pathophysiology. Evidence comes primarily from disease models and conclusive data to support bioenergetic dysfunction in vivo in patients is currently lacking. This study is the first to assess mitochondrial dysfunction in brain and muscle in individuals living with ALS using 31P-magnetic resonance spectroscopy (MRS), the modality of choice to assess energy metabolism in vivo. We recruited 20 patients and 10 healthy age and gender-matched control subjects in this cross-sectional clinico-radiological study. 31P-MRS was acquired from cerebral motor regions and from tibialis anterior during rest and exercise. Bioenergetic parameter estimates were derived including: ATP, phosphocreatine, inorganic phosphate, adenosine diphosphate, Gibbs free energy of ATP hydrolysis (ΔGATP), phosphomonoesters, phosphodiesters, pH, free magnesium concentration, and muscle dynamic recovery constants. Linear regression was used to test for associations between brain data and clinical parameters (revised amyotrophic functional rating scale, slow vital capacity, and upper motor neuron score) and between muscle data and clinico-neurophysiological measures (motor unit number and size indices, force of contraction, and speed of walking). Evidence for primary dysfunction of mitochondrial oxidative phosphorylation was detected in the brainstem where ΔGATP and phosphocreatine were reduced. Alterations were also detected in skeletal muscle in patients where resting inorganic phosphate, pH, and phosphomonoesters were increased, whereas resting ΔGATP, magnesium, and dynamic phosphocreatine to inorganic phosphate recovery were decreased. Phosphocreatine in brainstem correlated with respiratory dysfunction and disability; in muscle, energy metabolites correlated with motor unit number index, muscle power, and speed of walking. This study provides in vivo evidence for bioenergetic dysfunction in ALS in brain and skeletal muscle, which appears clinically and electrophysiologically relevant. 31P-MRS represents a promising technique to assess the pathophysiology of mitochondrial function in vivo in ALS and a potential tool for future clinical trials targeting bioenergetic dysfunction.

Knowledge and attitudes of critical care providers towards neurophysiological monitoring, seizure diagnosis, and treatment.

AIM: To explore the attitudes of paediatric intensive care unit (PICU) health care professionals towards diagnosis and neurophysiological monitoring of seizures. METHOD: This study used an explanatory sequential mixed-methods approach, interconnecting quantitative and qualitative features, comprising questionnaires and interviews, with equal weighting between stages, of health care professionals working in UK PICUs. Interview data were analysed using thematic analysis and triangulated with questionnaire data. RESULTS: Seventy-two questionnaires were returned: 49 out of 60 (71.0%) of respondents reported that seizures were extremely hard or somewhat hard to diagnose in a critically ill child, and 81.2% had seen misdiagnosis occur. Thematic analysis revealed two main themes: (1) feeling out of control when faced with 'grey areas'; and (2) regaining control, which compromised three subthemes: aggressive intervention, accurate diagnosis, and eschewing diagnosis. INTERPRETATION: Health care professionals find accurate diagnosis of seizures difficult, particularly in sedated/paralysed children and those with chronic neurological disorders. They report they would like better educational opportunities on discriminating between epileptic and non-epileptic events to improve their confidence. Professionals want routine neurophysiological monitoring that can be applied and interpreted at the bedside throughout the day to regain a sense of control over their patient, direct treatment appropriately, and, potentially, improve outcomes, but report appropriate training and peer review are essential if it is to be introduced into routine care. What this study adds Paediatric intensive care unit (PICU) staff feel out of control when faced with diagnosing seizures. Neurophysiological monitoring is wanted to help diagnosis and treatment. Amplitude-integrated electroencephalography is the preferred, pragmatic tool by PICU staff.

Investigating and managing neonatal seizures in the UK: an explanatory sequential mixed methods approach.

BACKGROUND: Neonatal seizures are difficult to diagnose and, when they are, tradition dictates first line treatment is phenobarbital. There is little data on how consultants diagnose neonatal seizures, choose when to treat or how they choose aetiological investigations or drug treatments. The purpose of this study was to assess the variation across the UK in the management of neonatal seizures and explore paediatricians' views on their diagnosis and treatment. METHODS: An explanatory sequential mixed methods approach was used (QUAN→QUAL) with equal waiting between stages. We collected quantitative data from neonatology staff and paediatric neurologists using a questionnaire sent to neonatal units and via emails from the British Paediatric Neurology Association. We asked for copies of neonatal unit guidelines on the management of seizures. The data from questionnaires was used to identify16 consultants using semi-structured interviews. Thematic analysis was used to interpret qualitative data, which was triangulated with quantitative questionnaire data. RESULTS: One hundred questionnaires were returned: 47.7% thought levetiracetam was as, or equally, effective as phenobarbital; 9.2% thought it was less effective. 79.6% of clinicians had seen no side effects in neonates with levetiracetam. 97.8% of unit guidelines recommended phenobarbital first line, with wide variation in subsequent drug choice, aetiological investigations, and advice on when to start treatment. Thematic analysis revealed three themes: 'Managing uncertainty with neonatal seizures', 'Moving practice forward' and 'Multidisciplinary team working'. Consultants noted collecting evidence on anti-convulsant drugs in neonates is problematic, and recommended a number of solutions, including collaboration to reach consensus guidelines, to reduce diagnostic and management uncertainty. CONCLUSIONS: There is wide variation in the management of neonatal seizures and clinicians face many uncertainties. Our data has helped reveal some of the reasons for current practice and decision making. Suggestions to improve certainty include: educational initiatives to improve the ability of neonatal staff to describe suspicious events, greater use of video, closer working between neonatologists and neurologists, further research, and a national discussion to reach a consensus on a standardised approach to managing neonatal epileptic seizures.

Sleep deprivation and melatonin for inducing sleep in paediatric electroencephalography: a prospective multicentre service evaluation.

AIM: To compare the efficacy of the main methodologies in attaining sleep and electroencephalography (EEG) abnormalities in children with a view to producing recommendations on best practice. METHOD: Fifty-one UK centres participated. Methods for sleep induction (sleep deprivation, melatonin, and combined sleep deprivation/melatonin) were compared. Data pertaining to demographics, achievement of stage II sleep, and recording characteristics (duration of study, presence of epileptiform activity in awake/sleep states) were prospectively collected for consecutive patients in November and December 2013. RESULTS: Five hundred and sixty-five patients were included. Age range was 1 years to 17 years (mean 7y 10mo), 27.7 per cent had an underlying neurobehavioural condition. Stage II sleep was achieved in 69 per cent of sleep deprived studies, 77 per cent of melatonin studies, and 90 per cent of combined intervention studies (p<0.001, χ2 ). In children who slept, there was no difference between the three interventions in eliciting epileptiform discharges. In children who did not sleep, epileptiform abnormalities were seen more often than after sleep deprivation alone (p=0.02, χ2 ). Seizures were rare. INTERPRETATION: Combined sleep deprivation/melatonin is more effective than either method alone in achieving sleep. The occurrence of epileptiform activity during sleep is broadly similar across the three groups. We recommend the combined intervention to induce sleep for paediatric EEG. WHAT THIS PAPER ADDS: Combined sleep deprivation/melatonin is more effective in achieving sleep than either sleep deprivation or melatonin alone. Sleep latency is shorter with combined sleep deprivation/melatonin. When children do sleep, there is no difference in the occurrence of epileptiform abnormalities between different induction methods. Seizures are rare in sleep electroencephalography recordings.

Imaging muscle as a potential biomarker of denervation in motor neuron disease.

OBJECTIVE: To assess clinical, electrophysiological and whole-body muscle MRI measurements of progression in patients with motor neuron disease (MND), as tools for future clinical trials, and to probe pathophysiological mechanisms in vivo. METHODS: A prospective, longitudinal, observational, clinicoelectrophysiological and radiological cohort study was performed. Twenty-nine patients with MND and 22 age-matched and gender-matched healthy controls were assessed with clinical measures, electrophysiological motor unit number index (MUNIX) and T2-weighted whole-body muscle MRI, at first clinical presentation and 4 months later. Between-group differences and associations were assessed using age-adjusted and gender-adjusted multivariable regression models. Within-subject longitudinal changes were assessed using paired t-tests. Patterns of disease spread were modelled using mixed-effects multivariable regression, assessing associations between muscle relative T2 signal and anatomical adjacency to site of clinical onset. RESULTS: Patients with MND had 30% higher relative T2 muscle signal than controls at baseline (all regions mean, 95% CI 15% to 45%, p<0.001). Higher T2 signal was associated with greater overall disability (coefficient -0.009, 95% CI -0.017 to -0.001, p=0.023) and with clinical weakness and lower MUNIX in multiple individual muscles. Relative T2 signal in bilateral tibialis anterior increased over 4 months in patients with MND (right: 10.2%, 95% CI 2.0% to 18.4%, p=0.017; left: 14.1%, 95% CI 3.4% to 24.9%, p=0.013). Anatomically, contiguous disease spread on MRI was not apparent in this model. CONCLUSIONS: Whole-body muscle MRI offers a new approach to objective assessment of denervation over short timescales in MND and enables investigation of patterns of disease spread in vivo. Muscles inaccessible to conventional clinical and electrophysiological assessment may be investigated using this methodology.

The role of cranial and thoracic electromyography within diagnostic criteria for amyotrophic lateral sclerosis.

INTRODUCTION: The contribution of cranial and thoracic region electromyography (EMG) to diagnostic criteria for amyotrophic lateral sclerosis (ALS) has not been evaluated. METHODS: Clinical and EMG data from each craniospinal region were retrospectively assessed in 470 patients; 214 had ALS. Changes to diagnostic classification in Awaji-Shima and revised El Escorial criteria after withdrawal of cranial/thoracic EMG data were ascertained. RESULTS: Sensitivity for lower motor neuron involvement in ALS was highest in the cervical/lumbar regions; specificity was highest in cranial/thoracic regions. Cranial EMG contributed to definite/probable Awaji-Shima categorization in 1.4% of patients. Thoracic EMG made no contribution. For revised El Escorial criteria, cranial and thoracic data reclassified 1% and 5% of patients, respectively. CONCLUSION: Cranial EMG data make small contributions to both criteria, whereas thoracic data contribute only to the revised El Escorial criteria. However, cranial and thoracic region abnormalities are specific in ALS. Consideration should be given to allowing greater diagnostic contribution from thoracic EMG. Muscle Nerve 54: 378-385, 2016.

Sensory Ganglionopathy and the Blink Reflex: Electrophysiological Features.

BACKGROUND: Sensory ganglionopathy (SG) is characterised by asymmetrical sensory fibre degeneration, with the primary pathology occurring at the level of the dorsal root ganglion. It is seen in the context of autoimmune, paraneoplastic, and degenerative disorders. There is limited literature examining the electrophysiological correlate of the trigeminal ganglion and associated pathways, the blink reflex (BR), in cases of SG. Previous work has suggested that the BR is preserved in cases of SG associated with paraneoplasia. METHODS: The local clinical neurophysiology database was searched for patients diagnosed with SG from peripheral nerve conduction studies in whom the BR was performed. Twenty-six patients were included in the final analysis. RESULTS: Sjögren's syndrome constituted the most common SG aetiology (8/26), followed by idiopathic cases (7/26) and paraneoplasia (5/26). BR abnormalities were seen in 9 of the 26 patients (34.6%) across all aetiologies. No patients reported sensory disturbance in the distribution of the trigeminal nerve, indicating that the changes noted are subclinical. Three patients showed abnormality of the R1 response; in the remaining six patients, only R2 responses were affected. CONCLUSIONS: Subclinical abnormalities of both R1 and R2 can be seen in the context of SG of varying aetiologies, including paraneoplasia. Performing the BR in patients with suspected of having SG may be helpful in providing additional evidence of patchy sensory fibre involvement that is characteristic of the disease.

Sensory ganglionopathy with livedoid vasculopathy controlled by immunotherapy.

INTRODUCTION: Livedoid vasculopathy is a rare dermatological condition characterized by painful ulceration, atrophic scarring, and persistent livedo reticularis. The pathogenesis is unclear. METHODS: We report a patient with biopsy-proven livedoid vasculopathy who developed a progressive sensory ganglionopathy with profound sensory ataxia. Serial nerve conduction assessments were undertaken. RESULTS: Combined treatment with prednisolone and mycophenolate mofetil failed to control the ganglionopathy. After addition of rituximab, both symptoms and nerve conduction studies showed stabilization. CONCLUSIONS: Sensory ganglionopathies associated with autoimmune and inflammatory conditions may be characterized by a sub-population of "sick" dorsal root ganglia that can be rescued with aggressive immunotherapy.

Neonatal seizures-part 2: Aetiology of acute symptomatic seizures, treatments and the neonatal epilepsy syndromes.

Most neonatal epileptic seizures are provoked by an underlying condition or problem-'acute symptomatic seizures'. However, a few neonatal epilepsy syndromes exist, and these are defined by the constellation of seizure types, EEG findings and family history seen. Making an accurate diagnosis of an epilepsy syndrome can help direct investigations, treatment options and provide prognostic information. This article discusses the investigative approach and treatments for neonatal epileptic seizures, including the neonatal epilepsy syndromes.

Neonatal seizures-part 1: Not everything that jerks, stiffens and shakes is a fit.

The neonatal period is the most frequent time of life to have epileptic seizures. However, neonates can also exhibit unusual movements that are not epileptic seizures. Differentiating between epileptic and non-epileptic movements can be difficult. Many neonatal seizures exhibit few or no clinical features at all. This article is for the benefit of paediatric trainees and reviews the published evidence on which neonatal movements are likely to be epileptic seizures and which are not. We also discuss epileptic seizure classification.

Contribution of neurophysiology to the diagnosis and monitoring of ALS.

This chapter describes the role of neurophysiological techniques in diagnosing and monitoring amyotrophic lateral sclerosis (ALS). Despite many advances, electromyography (EMG) remains a keystone investigation from which to build support for a diagnosis of ALS, demonstrating the pathophysiological processes of motor unit hyperexcitability, denervation and reinnervation. We consider development of the different diagnostic criteria and the role of EMG therein. While not formally recognised by established diagnostic criteria, we discuss the pioneering studies that have demonstrated the diagnostic potential of transcranial magnetic stimulation (TMS) of the motor cortex and highlight the growing evidence for TMS in the diagnostic process. Finally, accurately monitoring disease progression is crucial for the successful implementation of clinical trials. Neurophysiological measures of disease state have been incorporated into clinical trials for over 20 years and we review prominent techniques for assessing disease progression.

predicTTE: An accessible and optimal tool for time-to-event prediction in neurological diseases.

Time-to-event prediction is a key task for biological discovery, experimental medicine, and clinical care. This is particularly true for neurological diseases where development of reliable biomarkers is often limited by difficulty visualising and sampling relevant cell and molecular pathobiology. To date, much work has relied on Cox regression because of ease-of-use, despite evidence that this model includes incorrect assumptions. We have implemented a set of deep learning and spline models for time-to-event modelling within a fully customizable 'app' and accompanying online portal, both of which can be used for any time-to-event analysis in any disease by a non-expert user. Our online portal includes capacity for end-users including patients, Neurology clinicians, and researchers, to access and perform predictions using a trained model, and to contribute new data for model improvement, all within a data-secure environment. We demonstrate a pipeline for use of our app with three use-cases including imputation of missing data, hyperparameter tuning, model training and independent validation. We show that predictions are optimal for use in downstream applications such as genetic discovery, biomarker interpretation, and personalised choice of medication. We demonstrate the efficiency of an ensemble configuration, including focused training of a deep learning model. We have optimised a pipeline for imputation of missing data in combination with time-to-event prediction models. Overall, we provide a powerful and accessible tool to develop, access and share time-to-event prediction models; all software and tutorials are available at www.predictte.org.