Full screen

Share

Show pages

Siqi Li
Department of Biological Science
School of Life Sciences
University of Liverpool
Death Risk Prediction in Patients with Heart Failure
Want to make interactive content? It’s easy in Genially!

Over 30 million people build interactive content in Genially.

Check out what others have designed:

Transcript

Siqi Li Department of Biological ScienceSchool of Life SciencesUniversity of Liverpool

Death Risk Prediction in Patients with Heart Failure

  • Conclusion
  • Data Analysis
  • Hypothesis

index

  • Introduction

Development of indicator that can help to predict death risks by using the collected database.

Mainly focused on the relationship between ejection fraction and Heart failue.

Source: http://www.hrudved.com/ejection-fraction.asp

Current Study

Source: https://machinelearning-blog.com/2017/11/19/fsgdhfju/

Machine Learning

Introduction

  • Albumin
  • Creatinine
  • CPK enzyme
  • Glucose
  • Na+
  • Ca2+
  • Cholesterol
  • ……

Serum Biochemical Index

Potential and predictive value of Serum Biochemical Index in survival evaluation among patients with the diagnosis of heart failure

Hypothesis

Table 1. Meanings, measurement units, and intervals of each feature of the dataset

Davide Chicco and Giuseppe Jurman (2020), BMC Medical Informatics & Decision Making, 2/3/2020, Vol. 20 Issue 1, p1-16, 16p, 11 Charts, 4 Graphs, Chart; found on p3, https://doi.org/10.1186/s12911-020-1023-5

Data analysis

  • Level of Creatinine Phosphokinase
  • Level of Serum creatinine
  • Level of Serum sodium

Numerical data

  • Diagnosis of Diabetes
( in terms of serum insulin and glucose)

Boolean data

Figure 1. A Bar charts for categorical data consisting of death event with statistical details about the percentage of diagnostic of diabetes included in the plot of each bar.

Table 2. A two-way contingency table summarized the frequency distribution of death events during the followed-up period associated with the diagnosis of diabetes.

Boolean Data analysis

Table 3. Statistical analysis on numerical data with Mann-Whitney to test if a significant difference emerged.

Numerical Data analysis

  • Small sample
  • Limited followed-up period
  • Possibly inaccurate recordings
  • Great importance for research
  • Great value for reference.
  • Wider applied in the futrue
  • Merely the level of serum creatinine and sodium in the blood shown significant difference in death events during followed peirod
  • The level of serum creatinine and sodium are possible to be used as prognostication for predict death event in heart failure
  • The level of CPK enzyme could not be a significant indicator.
  • the diagnosis of diabetes could not serve as a diagnostic tool for mortality of heart failure

Conclusion

03

01

04

02

Anticipation of the Machine learning

Limitation of the given dataset

Value of Serum Chemical Index

Conclusion based on the given dataset

  • Davide, C. and Giuseppe, J. 2020. Machine learning can predict survival of patients with heart failure from serum creatinine and ejection fraction alone. BMC Medical Informatics and Decision Making, 20 (1), 1-16.
  • Helmenstine, A.M., Ph.D. (2019) "List of Common Blood Chemistry Tests." ThoughtCo. Available from: https://www.thoughtco.com/common-blood-chemistry-tests-608417 [Accessed 28th Feb, 2021].
  • Kent, M. (2007) ‘blood chemistry tests’, The Oxford Dictionary of Sports Science & Medicine. doi: 10.1093/acref/9780198568506.013.0923.
  • Lewis, G. A., Schelbert, E. B., Williams, S. G., Cunnington, C., Ahmed, F., et al. 2017. Biological Phenotypes of Heart Failure With Preserved Ejection Fraction. Journal of the American College of Cardiology, 70 (17), 2186-2200.
  • Nanayakkara, S. and Kaye, D. M. 2017. Targets for Heart Failure With Preserved Ejection Fraction. Clin Pharmacol Ther, 102 (2), 228-237.
  • Sakamoto, M., Fukuda, H., Kim, J., Ide, T., Kinugawa, S., et al. 2018. The impact of creating mathematical formula to predict cardiovascular events in patients with heart failure. Sci Rep, 8 (1), 3986.
  • Shilaskar, S. and Ghatol, A. 2013. Feature selection for medical diagnosis : Evaluation for cardiovascular diseases. Expert Systems with Applications, 40 (10), 4146-4153.
  • Weng, S. F., Reps, J., Kai, J., Garibaldi, J. M. and Qureshi, N. 2017. Can machine-learning improve cardiovascular risk prediction using routine clinical data? PLOS ONE, 12 (4), e0174944.

Reference

thanks

Q&A

Next page

genially options