Machine Learning for Cancer Diagnosis and Treatment
We explore the power of machine learning to revolutionize the analysis of Apparent Diffusion Coefficient (ADC) values from diffusion-weighted MRI scans. Our goal is to improve the accuracy of colorectal cancer diagnosis, tailor treatment plans, and ultimately enhance patient outcomes through innovative data analysis techniques.
Objectives
Personalize Treatment Plans: Based on their unique ADC profiles, predict the most effective treatment strategies for individual colorectal cancer patients.
Streamline Clinical Workflows: Integrate machine learning models into existing medical practices to save time and improve efficiency.
Enhance Clinical Decision-Making: Provide healthcare professionals with advanced tools to make informed decisions about colorectal cancer patient care.
Improve Diagnostic Accuracy: More precise interpretations of ADC values, reduced variability, and improved reliability of colorectal cancer diagnoses.
Number of MRI Scans in Romania
Our target for finishing the first stage of the research project is to apply it in real-world scenarios around clinics in Romania
958 520
MRI Scans between 2019-2021
Data acquired from Eurostat, online data code: hlth_co_exam, DOI:https://doi.org/10.2908/HLTH_CO_EXAM
Benefits
Read More
- For Surgeons and Radiologists: Enhanced tools for planning and decision-making, leading to better surgical outcomes and reduced risks in colorectal cancer treatment.
- For Oncologists: More accurate predictions of treatment responses aid in selecting the most effective therapies for colorectal cancer.
- For Patients: Personalized care plans that increase the likelihood of successful treatment and improve the quality of life for colorectal cancer patients.
Features
Read More
High Precision Analysis: Advanced algorithms provide detailed and accurate ADC interpretations for colorectal cancer.
Predictive Insights: Our models predict treatment efficacy, helping to tailor personalized treatment plans for colorectal cancer patients.
User-Friendly Interface: Designed for easy integration into clinical workflows, offering accessible and actionable insights for healthcare professionals.
Continuous Improvement: Our models learn from new data to continually improve accuracy and predictive power for colorectal cancer treatment.
Methodology
Read More
Data Collection: We gather retrospective diffusion-weighted MRI scans with multiple b-values, patient outcomes, and clinical histories, specifically focusing on colorectal cancer cases.
ADC Calculation: Calculate ADC maps from the DWI scans at different b-values for comprehensive analysis of colorectal tumors.
Machine Learning Application: Train machine learning models to correlate ADC values with clinical outcomes in colorectal cancer, identifying the most predictive b-values.
Outcome Prediction: Use the models to predict treatment responses and guide clinical decisions, aiming to optimize colorectal cancer patient care.
Team
Alex-Edward Debelka is a fifth-year medical student at the University of Oradea, Romania, and a three-time Erasmus student at Sorbonne University, Paris, focusing on surgical oncology and AI-driven diagnostics.
Daniele Moș is an Erasmus Master’s student at TU Dresden from Cluj-Napoca, Romania, specializing in machine learning and AI-guided surgery.
Dr. Adrian Coțe is a general surgeon at the Emergency County Hospital of Oradea. His doctoral thesis, “Optimization of Diagnostic and Treatment Methods in Rectal Cancer,” focuses on implementing ADC values to enhance diagnostic accuracy and treatment efficacy.
Transforming Tomorrow’s Healthcare with Today’s Innovation. Join Us in Pioneering Precision Oncology.
Get in Touch with Us
alex@cyapiens.com
+40726081999
daniele@cyapiens.com
Mimozei 10
Oradea, Bihor, Romania