Several models have been developed to predict prolonged stay in the intensive care unit (ICU) after cardiac surgery. However, no extensive quantitative validation of these models has yet been conducted. This study sought to identify and validate existing prediction models for prolonged ICU length of stay after cardiac surgery.
Data-driven condition-based maintenance (CBM) and predictive maintenance (PdM) strategies have emerged over recent years and aim at minimizing the aviation maintenance costs and environmental impact by the diagnosis and prognosis of aircraft systems. As the use of data and relevant algorithms is essential to AI-based gas turbine diagnostics, there are different technical, operational, and regulatory challenges that need to be tackled in order for the aeronautical industry to be able to exploit their full potential. In this work, the machine learning (ML) method of the generalised additive model (GAM) is used in order to predict the evolution of an aero engine’s exhaust gas temperature (EGT). Three different continuous synthetic data sets developed by NASA are employed, known as New Commercial Modular Aero-Propulsion System Simulation (N-CMAPSS), with increasing complexity in engine deterioration. The results show that the GAM can be predict the evolution of the EGT with high accuracy when using several input features that resemble the types of physical sensors installed in aero gas turbines currently in operation. As the GAM offers good interpretability, this case study is used to discuss the different data attributes a data set needs to have in order to build trust and move towards certifiable models in the future.
OBJECTIVE: To further test the validity and clinical usefulness of the steep ramp test (SRT) in estimating exercise tolerance in cancer survivors by external validation and extension of previously published prediction models for peak oxygen consumption (Vo2peak) and peak power output (Wpeak).DESIGN: Cross-sectional study.SETTING: Multicenter.PARTICIPANTS: Cancer survivors (N=283) in 2 randomized controlled exercise trials.INTERVENTIONS: Not applicable.MAIN OUTCOME MEASURES: Prediction model accuracy was assessed by intraclass correlation coefficients (ICCs) and limits of agreement (LOA). Multiple linear regression was used for model extension. Clinical performance was judged by the percentage of accurate endurance exercise prescriptions.RESULTS: ICCs of SRT-predicted Vo2peak and Wpeak with these values as obtained by the cardiopulmonary exercise test were .61 and .73, respectively, using the previously published prediction models. 95% LOA were ±705mL/min with a bias of 190mL/min for Vo2peak and ±59W with a bias of 5W for Wpeak. Modest improvements were obtained by adding body weight and sex to the regression equation for the prediction of Vo2peak (ICC, .73; 95% LOA, ±608mL/min) and by adding age, height, and sex for the prediction of Wpeak (ICC, .81; 95% LOA, ±48W). Accuracy of endurance exercise prescription improved from 57% accurate prescriptions to 68% accurate prescriptions with the new prediction model for Wpeak.CONCLUSIONS: Predictions of Vo2peak and Wpeak based on the SRT are adequate at the group level, but insufficiently accurate in individual patients. The multivariable prediction model for Wpeak can be used cautiously (eg, supplemented with a Borg score) to aid endurance exercise prescription.
Huntington’s disease (HD) and various spinocerebellar ataxias (SCA) are autosomal dominantly inherited neurodegenerative disorders caused by a CAG repeat expansion in the disease-related gene1. The impact of HD and SCA on families and individuals is enormous and far reaching, as patients typically display first symptoms during midlife. HD is characterized by unwanted choreatic movements, behavioral and psychiatric disturbances and dementia. SCAs are mainly characterized by ataxia but also other symptoms including cognitive deficits, similarly affecting quality of life and leading to disability. These problems worsen as the disease progresses and affected individuals are no longer able to work, drive, or care for themselves. It places an enormous burden on their family and caregivers, and patients will require intensive nursing home care when disease progresses, and lifespan is reduced. Although the clinical and pathological phenotypes are distinct for each CAG repeat expansion disorder, it is thought that similar molecular mechanisms underlie the effect of expanded CAG repeats in different genes. The predicted Age of Onset (AO) for both HD, SCA1 and SCA3 (and 5 other CAG-repeat diseases) is based on the polyQ expansion, but the CAG/polyQ determines the AO only for 50% (see figure below). A large variety on AO is observed, especially for the most common range between 40 and 50 repeats11,12. Large differences in onset, especially in the range 40-50 CAGs not only imply that current individual predictions for AO are imprecise (affecting important life decisions that patients need to make and also hampering assessment of potential onset-delaying intervention) but also do offer optimism that (patient-related) factors exist that can delay the onset of disease.To address both items, we need to generate a better model, based on patient-derived cells that generates parameters that not only mirror the CAG-repeat length dependency of these diseases, but that also better predicts inter-patient variations in disease susceptibility and effectiveness of interventions. Hereto, we will use a staggered project design as explained in 5.1, in which we first will determine which cellular and molecular determinants (referred to as landscapes) in isogenic iPSC models are associated with increased CAG repeat lengths using deep-learning algorithms (DLA) (WP1). Hereto, we will use a well characterized control cell line in which we modify the CAG repeat length in the endogenous ataxin-1, Ataxin-3 and Huntingtin gene from wildtype Q repeats to intermediate to adult onset and juvenile polyQ repeats. We will next expand the model with cells from the 3 (SCA1, SCA3, and HD) existing and new cohorts of early-onset, adult-onset and late-onset/intermediate repeat patients for which, besides accurate AO information, also clinical parameters (MRI scans, liquor markers etc) will be (made) available. This will be used for validation and to fine-tune the molecular landscapes (again using DLA) towards the best prediction of individual patient related clinical markers and AO (WP3). The same models and (most relevant) landscapes will also be used for evaluations of novel mutant protein lowering strategies as will emerge from WP4.This overall development process of landscape prediction is an iterative process that involves (a) data processing (WP5) (b) unsupervised data exploration and dimensionality reduction to find patterns in data and create “labels” for similarity and (c) development of data supervised Deep Learning (DL) models for landscape prediction based on the labels from previous step. Each iteration starts with data that is generated and deployed according to FAIR principles, and the developed deep learning system will be instrumental to connect these WPs. Insights in algorithm sensitivity from the predictive models will form the basis for discussion with field experts on the distinction and phenotypic consequences. While full development of accurate diagnostics might go beyond the timespan of the 5 year project, ideally our final landscapes can be used for new genetic counselling: when somebody is positive for the gene, can we use his/her cells, feed it into the generated cell-based model and better predict the AO and severity? While this will answer questions from clinicians and patient communities, it will also generate new ones, which is why we will study the ethical implications of such improved diagnostics in advance (WP6).
In order to stay competitive and respond to the increasing demand for steady and predictable aircraft turnaround times, process optimization has been identified by Maintenance, Repair and Overhaul (MRO) SMEs in the aviation industry as their key element for innovation. Indeed, MRO SMEs have always been looking for options to organize their work as efficient as possible, which often resulted in applying lean business organization solutions. However, their aircraft maintenance processes stay characterized by unpredictable process times and material requirements. Lean business methodologies are unable to change this fact. This problem is often compensated by large buffers in terms of time, personnel and parts, leading to a relatively expensive and inefficient process. To tackle this problem of unpredictability, MRO SMEs want to explore the possibilities of data mining: the exploration and analysis of large quantities of their own historical maintenance data, with the meaning of discovering useful knowledge from seemingly unrelated data. Ideally, it will help predict failures in the maintenance process and thus better anticipate repair times and material requirements. With this, MRO SMEs face two challenges. First, the data they have available is often fragmented and non-transparent, while standardized data availability is a basic requirement for successful data analysis. Second, it is difficult to find meaningful patterns within these data sets because no operative system for data mining exists in the industry. This RAAK MKB project is initiated by the Aviation Academy of the Amsterdam University of Applied Sciences (Hogeschool van Amsterdan, hereinafter: HvA), in direct cooperation with the industry, to help MRO SMEs improve their maintenance process. Its main aim is to develop new knowledge of - and a method for - data mining. To do so, the current state of data presence within MRO SMEs is explored, mapped, categorized, cleaned and prepared. This will result in readable data sets that have predictive value for key elements of the maintenance process. Secondly, analysis principles are developed to interpret this data. These principles are translated into an easy-to-use data mining (IT)tool, helping MRO SMEs to predict their maintenance requirements in terms of costs and time, allowing them to adapt their maintenance process accordingly. In several case studies these products are tested and further improved. This is a resubmission of an earlier proposal dated October 2015 (3rd round) entitled ‘Data mining for MRO process optimization’ (number 2015-03-23M). We believe the merits of the proposal are substantial, and sufficient to be awarded a grant. The text of this submission is essentially unchanged from the previous proposal. Where text has been added – for clarification – this has been marked in yellow. Almost all of these new text parts are taken from our rebuttal (hoor en wederhoor), submitted in January 2016.
Every year in the Netherlands around 10.000 people are diagnosed with non-small cell lung cancer, commonly at advanced stages. In 1 to 2% of patients, a chromosomal translocation of the ROS1 gene drives oncogenesis. Since a few years, ROS1+ cancer can be treated effectively by targeted therapy with the tyrosine kinase inhibitor (TKI) crizotinib, which binds to the ROS1 protein, impairs the kinase activity and thereby inhibits tumor growth. Despite the successful treatment with crizotinib, most patients eventually show disease progression due to development of resistance. The available TKI-drugs for ROS1+ lung cancer make it possible to sequentially change medication as the disease progresses, but this is largely a ‘trial and error’ approach. Patients and their doctors ask for better prediction which TKI will work best after resistance occurs. The ROS1 patient foundation ‘Stichting Merels Wereld’ raises awareness and brings researchers together to close the knowledge gap on ROS1-driven oncogenesis and increase the options for treatment. As ROS1+ lung cancer is rare, research into resistance mechanisms and the availability of cell line models are limited. Medical Life Sciences & Diagnostics can help to improve treatment by developing new models which mimic the situation in resistant tumor cells. In the current proposal we will develop novel TKI-resistant cell lines that allow screening for improved personalized treatment with TKIs. Knowledge of specific mutations occurring after resistance will help to predict more accurately what the next step in patient treatment could be. This project is part of a long-term collaboration between the ROS1 patient foundation ‘Stichting Merels Wereld’, the departments of Pulmonary Oncology and Pathology of the UMCG and the Institute for Life Science & Technology of the Hanzehogeschool. The company Vivomicx will join our consortium, adding expertise on drug screening in complex cell systems.
