Skip to content
Learning

Colon cancer metastasis risk prediction simulator

Early detection and prognosis of colon cancer metastasis are critical for patient survival, yet remain challenging due to the interaction of clinical, pathological, and genetic factors. We present a machine learning simulator for colon cancer liver metastasis risk prediction that estimates 3-year mortality by integrating genomic, mutational, and clinical data, providing individualized risk scores to support oncologists and researchers.

This tool bridges clinical and genomic data with practical applications, helping improve prognosis assessment in metastatic colon cancer.

Healthcare professionals can test it with Neural Designer.

Download free trial

Inputs:

Type of Adenocarcinoma:
Metastasis primary site count:
Microsatellite instability score:
Microsatellite instability type:
Mutation count:
Cancer microsatellite subtype:
Tumour mutational burden:
KIT mutation count:
CARD11 mutation count:
RB1 mutation count:
WT1 mutation count:
PLCG2 mutation count:
DNMT1 mutation count:
BRD4 mutation count:
PIK3R1 mutation count:
IRS2 mutation count:
SESN1 mutation count:
NPM1 mutation count:
Liver metastasis risk:6.33%

Related posts

Liver metastasis gene mutationsMetastasis risk assessment using machine learning Liver metastasis gene mutationsPrimary tumor site classification Breast cancer geneBreast cancer mortality prediction with machine learning Pancreatic cancerPancreatic cancer prediction with machine learning

Scope and limitations

This simulator is an educational and research demonstration of a statistical model. Its output depends on the variables, population, data quality and evaluation procedure used to create the model. A prediction for an individual does not establish a diagnosis, prognosis or treatment recommendation.

Do not use the result for clinical decisions or to replace qualified medical assessment. Before any real-world use, an independently governed clinical workflow would need representative validation data, documented inclusion criteria, calibration checks, uncertainty analysis, monitoring and approval under the applicable requirements.

For the modelling workflow behind this type of example, review testing analysis and the medical prognosis use case.