Scribd
Upload
2 views

Detecting Thyroid Cancer Recurrence Using Patient Data

The document discusses the importance of early detection of thyroid cancer recurrence and its impact on treatment and patient outcomes. It explores the use of machine learning models for predicting recurrence, detailing data sources, preprocessing techniques, and feature selection. The presentation emphasizes the significance of accurate predictions and includes case studies demonstrating the effects of early versus late detection.

Uploaded by

Faza Ulfath
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
2 views

Detecting Thyroid Cancer Recurrence Using Patient Data

The document discusses the importance of early detection of thyroid cancer recurrence and its impact on treatment and patient outcomes. It explores the use of machine learning models for predicting recurrence, detailing data sources, preprocessing techniques, and feature selection. The presentation emphasizes the significance of accurate predictions and includes case studies demonstrating the effects of early versus late detection.

Uploaded by

Faza Ulfath
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 8

Detecting Thyroid

Cancer Recurrence
using Patient Data
Early detection of thyroid cancer recurrence is crucial. It
significantly impacts treatment options and patient outcomes. This
presentation explores using patient data. Machine learning models
can predict recurrence. We will cover data sources, preprocessing,
and feature selection.

by Faza Ulfath
Introduction: The Challenge of Recurrence
The Recurrence Problem Impact on Patients

Thyroid cancer recurrence poses a significant clinical Recurrence can lead to increased morbidity and
challenge. It requires continuous monitoring and timely mortality. It also affects the patient's quality of life.
intervention. Early detection helps.
Data Sources and
Preprocessing
Techniques
1 Comprehensive 2 Data Cleaning
Data
Handling missing values.
Patient history and lab Correcting
results. Imaging reports. inconsistencies. Data
Treatment details. normalization.

3 Feature Engineering
Creating relevant variables. Combining data points.
Feature Selection:
Identifying Key
Predictors

Genetic Biochemical Imaging


Markers Markers Features
Specific gene Thyroglobulin Tumor size and
mutations. levels. Indicate location. Critical
Predictive of disease activity. recurrence factors.
recurrence risk.
Machine Learning
Models for Recurrence
Prediction
Random Forest

Support Vector Machines

Neural Networks

Logistic Regression
Model Performance and Validation Metrics

Key metrics include accuracy and precision. Also recall and F1-score. These assess the model's reliability. They ensure accurate recurrence predictions.
Case Studies: Real-
World Examples
Case 1: Early Detection
Model predicted recurrence. Treatment was promptly adjusted.
The patient achieved remission.

Case 2: Late Detection


Recurrence was missed initially. Delayed intervention resulted.
A poorer outcome followed.
Conclusion: Improving Patient Outcomes
Early Detection 1
Enhanced monitoring. Proactive interventions.

2 Personalized Treatment
Tailored therapies. Based on risk profiles.

Improved Survival 3
Better patient outcomes. Enhanced life quality.

You might also like