Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

Machine Learning in ICU mortality prediction


Published on

Several models using scoring techniques like APACHE and SAPS for mortality prediction have been developed to assess severity of illness and predict mortality in intensive care units (ICUs) standardizing research and assessing performance of ICUs. Machine learning can be employed to build better suited models with locally available data

Published in: Data & Analytics
  • How is it possible? 80% accuracy? Sports picks directly from the insiders? ♣♣♣
    Are you sure you want to  Yes  No
    Your message goes here
  • Be the first to like this

Machine Learning in ICU mortality prediction

  1. 1. Data Science for Healthcare June 27, 2017 ICU Mortality Prediction Case Study
  2. 2. Page 2 © AlgoAnalytics All rights reserved Outline Our Healthcare Solutions ICU Mortality Prediction Case Study About AlgoAnalytics
  3. 3. Page 3 © AlgoAnalytics All rights reserved Aniruddha Pant CEO and Founder of AlgoAnalytics PhD, Control systems, University of California at Berkeley, USA 2001 • 20+ years in application of advanced mathematical techniques to academic and enterprise problems. • Experience in application of machine learning to various business problems. • Experience in financial markets trading; Indian as well as global markets. Highlights • Experience in cross-domain application of basic scientific process. • Research in areas ranging from biology to financial markets to military applications. • Close collaboration with premier educational institutes in India, USA & Europe. • Active involvement in startup ecosystem in India. Expertise • Vice President, Capital Metrics and Risk Solutions • Head of Analytics Competency Center, Persistent Systems • Scientist and Group Leader, Tata Consultancy Services Prior Experience • Work at the intersection of mathematics and other domains • Harness data to provide insight and solutions to our clients Analytics Consultancy • +30 data scientists with experience in mathematics and engineering • Team strengths include ability to deal with structured/ unstructured data, classical ML as well as deep learning using cutting edge methodologies Led by Aniruddha Pant • Develop advanced mathematical models or solutions for a wide range of industries: • Financial services, Retail, economics, healthcare, BFSI, telecom, … Expertise in Mathematics and Computer Science • Work closely with domain experts – either from the clients side or our own – to effectively model the problem to be solved Working with Domain Specialists About AlgoAnalytics
  4. 4. Page 4 © AlgoAnalytics All rights reserved AlgoAnalytics - One Stop AI Shop •We use structured data to design our predictive analytics solutions like churn, recommender system •We use techniques like clustering, Recurrent Neural Networks, Structured Data •We use text data analytics for designing solutions like sentiment analysis, news summarization and many more •We use techniques like natural language processing, word2vec, deep learning, TF-IDF Text Data •Image data is used for predicting existence of particular pathology, image recognition and many others •We use techniques like deep learning – convolutional neural network, artificial neural networks and technologies like TensorFlow Image Data •We use sound data to design factory solutions like air leakage detection, identification of empty and loaded strokes from press data, engine-compressor fault detection •We use techniques like deep learning Sound Data BFSI •Dormancy Analysis •Recommender System •Credit/Collection Score Retail •Churn Analysis •Recommender System •Image Analytics Healthcare •Medical Image Diagnostics •Work flow optimization •Cash flow forecasting Legal •Contracts Management •Structured Document decomposition •Document similarity in text analytics Internet of Things •Predictive in ovens •Air leakage detection •Engine/compressor fault detection Others •Algorithmic trading strategies •Risk sensing – network theory •Network failure model
  5. 5. Page 5 © AlgoAnalytics All rights reserved ICU Mortality Prediction Case Study
  6. 6. Page 6 © AlgoAnalytics All rights reserved ICU Mortality Prediction: Why Machine Learning A host of different scoring techniques have been employed for scoring patient mortality in ICUs despite the fact that certain patient factors cannot be taken into account and cannot always be employed on local data with the same results… However, with Machine learning one can develop local based models using local data and thus have better discriminating ability to distinguish between patients with high risk and those with low risk of mortality and thus act as valid benchmarking, and also for monitoring ICU performance
  7. 7. Page 7 © AlgoAnalytics All rights reserved ICU Mortality Prediction Possiblities and Results Bootstrap comparison between actual and predicted mortality Accuracy: 86.43% Sensitivity: 58.15% Specificity: 92.48% ➢ It was possible to build a model with local data with performance comparable with APACHE II ➢ The variables used were more readily available than those used in APACHE II model (comparable with SAPII model) ➢ This Accuracy is calculated using simple Random Forest Model. ➢ We can get better accuracy with more data and fine tuning the models. • Predictive performance of traditional severity of illness scores (APACHE II, SAPS II etc.) can be enhanced using local hospital electronic medical records • Personalized mortality prediction with the use of digital data with patient similarity metric • New predictive models can be built using local data
  8. 8. Page 8 © AlgoAnalytics All rights reserved Other Work Done in Healthcare Medical diagnostics – Detecting serious disorders or diseases through image analytics. We have developed solutions in diabetic retinopathy, brain MRI scan, sonography and others Cash Flow Forecasting – Forecasting of cash flows based on claims history, reimbursement analysis and potential denials to forecast cash Work Flow Optimization– Using historical data for staffing to reduce costs, Having the right clinician at right time at right place Efficient Use of Hospital Resources – Prevent bottlenecks in urgent care by analyzing patient flow during peak times Grant problem - Predict likelihood that a particular proposal will receive grant using text analytics
  9. 9. Page 9 © AlgoAnalytics All rights reserved Technology
  10. 10. Interested in knowing more? June 27, 2017 Contact: