Covid-19 Patient Outcome Predictors


Covid-19 – Days To Viral Clearance (median)

 

Days To Virus Clearance (Median)

See Predictive Model Report

Covid-19 – Analytics Reference Ranges – For This Predictor

Sample CSV (download & fill out with your patient data and upload)

  • This is a clinician-grade tool. Use it to get an immediate insight!
  • As policy makers and healthcare professionals tackle the COVID-19 pandemic, a critical factor inhibiting effective decision making at regional, national, and global levels is a lack of relevant data on patient outcomes.
  • It’s based on: 
    • Dataset largely derived from studies run in hospitals and nations affected with COVID-19.
    • BIG NO-NO! The raw data that we used should not and was not used by us to estimate trends in the general population such as mortality rates.
    • It should be possible to derive reasonably accurate estimates for e.g. “Days to Viral Clearance (Median)” and it should be noted to take into account
      • (a) for the prevalence of asymptomatic patients, and
      • (b) including only sufficiently representative studies., which we have done in the model.
      • The dataset that we used for our AI deep-learning predictive model, aggregates data from over 160 published clinical studies and preprints released between December 2019 and April 2020.
      • We’ll keep on updating this resource as it’s updated from the source as more research investigations from other countries are getting published.
  • Data Credit: MIT Operations Research Center
  • Other important calculators (calcs) are given below!

Covid-19 Infection Risk Calculator

Your Infection Risk Calculator

  • Personalized calculator predicting results of COVID test.
  • This is a clinician-grade tool. Use it to get an immediate insight!

Covid-19 Mortality Risk Calculator
Your Mortality Risk Calculator
  • Personalized calculator predicting mortality upon hospitalization.
  • This is a clinician-grade tool. Use it to get an immediate insight!

Covid-19 Case Predictions
Case Predictions (see any country)
  • Epidemiological predictions of COVID-19 infections, hospital stays, and mortalities.
  • This is a clinician-grade tool. Use it to get an immediate insight!

Covid-19 Policy Evaluations
Policy Evaluations
  • Predicting infections and deaths based on various policy implementations.
  • This is a clinician-grade tool. Use it to get an immediate insight!

Ventilator Allocation
Ventilator Allocation
  • Leveraging delays between state peaks to optimally re-use ventilators.
  • This is a clinician-grade tool. Use it to get an immediate insight!

Covid-19 Analytics-Driven Financial Relief Planning
Financial Relief Planning
  • Unemployment is rising. How analytics can help inform COVID-19 related financial decisions.
  • This is a clinician-grade tool. Use it to get an immediate insight!

 


Risk Prediction: 4C Mortality Score
  • This is a clinician-grade tool. Use it to get immediate insights!
  • It’s Used to Determine In-House Mortality For Covid-19 Patient
  • Questions should be asked to determine effectiveness of treatment 
    • how long are people infectious, and what body fluids are infectious?
    • what puts people at higher risk of severe illness?
    • what is the best way to diagnose the disease?
    • who should we treat early with drugs, and which drugs cause harm?
    • does the immune system in some patients do more harm than good?
    • what other infections(such as pneumonia or flu) happen at the same time?

 

  • Admit to ICU predictor
    • This is a clinician-grade tool. Use it to get an immediate insight!
    • Using Individual patient data & presenting symptoms at hospital admission to quickly assess if ICU admission is required or not (Yes/No) 
    • A sample csv file format to be uploaded with the filled-in data for the predictions, can be seen in the app front-end.
    • Result: Your Predictions instantly
    • Use the above predictive engines to benefit patients by decreasing suffering and increase your site’s best-practices empirically & immediately.
    • Notes on CSV
      1. Mechanical Ventilation
      2. Limited to patients on mechanical ventilation
      3. Prone on mechanical ventilation within first 7 days of mechanical ventilation
      4. Use of vasopressor medications at any time during hospitalization

Covid-19 ICU Admit Predictor

See Predictive Model Report

Sample CSV (download & fill out with your patient data and upload)


  • DNR action for adverse event (AE) of death predictor
    • This is a clinician-grade tool. Use it to get an immediate insight!
    • Based on the possibility of the unfortunate eventual ultimate adverse-event (AE) , i.e. death, based on all available data.
    • Using 3-5 days of lab & ICU individual patient data, ARDS strategies, actual vasopressor use, etc.
    • Meaning, we can use our predictive neural network (neuralnet) by uploading a simple csv to predict on a quantitative-basis of chance of death as YES or NO to take action:
      • for humane do not resuscitate DNR choices for families, attending physicians and medical support staff. 
    • NIH DNR definition and order
    • A sample csv file format to be uploaded with the filled-in data for the predictions, can be seen in the app front-end.
    • Result: Your Predictions instantly
    • Use the above predictive engines to benefit patients by decreasing suffering and increase your site’s best-practices empirically & immediately.
    • Notes on CSV
      1. Mechanical Ventilation
      2. Limited to patients on mechanical ventilation
      3. Prone on mechanical ventilation within first 7 days of mechanical ventilation
      4. Use of vasopressor medications at any time during hospitalization

Covid-19 ICU DNR Predictor

Please Note: AIMLDL model is undergoing update – please visit later. Thank you.


References
  • NEJM Source: Covid-19 in Critically Ill Patients in the Seattle Region — Case Series
  • Arnold RM. Palliative care. In: Goldman L, Schafer AI, eds. Goldman-Cecil Medicine. 26th ed. Philadelphia, PA: Elsevier; 2020:chap 3.
  • Bullard MK. Medical ethics. In: Harken AH, Moore EE, eds. Abernathy’s Surgical Secrets. 7th ed. Philadelphia, PA: Elsevier; 2018:chap 106.
  • Moreno JD, DeKosky ST. Ethical considerations in the care of patients with neurosurgical disease. In: Cottrell JE, Patel P, eds. Cottrell and Patel’s Neuroanesthesia. 6th ed. Philadelphia, PA: Elsevier; 2017:chap 26.
Notes
  • Robust patient datasets. Data additions are done frequently to train and test to get more robust models built from them.
Review Date 10/22/2020 & Updated By

David C. Dugdale, III, MD, Professor of Medicine, Division of General Medicine, Department of Medicine, University of Washington School of Medicine, Seattle, WA. Also reviewed by David Zieve, MD, MHA, Medical Director, Brenda Conaway, Editorial Director, and the A.D.A.M. Editorial team.