Using Artificial Intelligence to Predict Rectal Cancer Outcomes

Sponsor

Taichung Veterans General Hospital (Other)

Overall Status

Completed

CT.gov ID

NCT05723965

Collaborator

(none)

720

Enrollment

Location

147

Actual Duration (Months)

4.9

Patients Per Site Per Month

Study Details

Study Description

Brief Summary

Investigator retrospective collect cases during 2010-2021 diagnosed as rectal adenocarcinoma with high quality CT images. Local advanced rectal cancer cases were labeled as "disease". Nor were defined " normal".

Using artificial intelligence CNN on jupyter notebook with open phyton code to train and develop models capable to recognizing local advanced rectal cancer. Modify the phyton code for better predict rate and help physician to quickly evaluate disease severity for fresh rectal cancer cases.

Condition or Disease	Intervention/Treatment	Phase
Rectal Cancer Stage III	Other: As training material for deep learning model. Other: As materials for external validation for the buildup model.

Detailed Description

From 2010.10.1~2021.12.31, rectal cancer patients with cT3-4 lesion was included. Collect high quality CT images with DICOM files in tumor segment. cT1-2, low rectal lesions, non-CRC cases were not included. Non-contrast and artificial defect images were also excluded. CT images were labeled as" diseased " when CRM were threatened (<2mm). All images were labeled according to judgment of 2 specialist. The data were separated into 2 parts. One for AI model training and testing, another for external validation. The training testing dataset was achieved by deep learning neural network and evaluating model accuracy performance. Then the model was applied into external validation dataset for real-world testing, evaluating coherent rate between AI and the Dr. decision. Furthermore, to see the cancer survival outcomes according to AI model prediction results.

Study Design

Study Type:

Observational

Actual Enrollment :

720 participants

Observational Model:

Case-Control

Time Perspective:

Retrospective

Official Title:

Using CNN Image Recognition to Predict Rectal Cancer Outcomes

Actual Study Start Date :

Oct 1, 2010

Actual Primary Completion Date :

Jul 31, 2022

Actual Study Completion Date :

Dec 31, 2022

Arms and Interventions

Arm	Intervention/Treatment
rectal cancer lesion images for training Rectal cancer lesion images. Images with threatened (<2mm) circumferential margin of rectal cancer were labeled as "diseased". Otherwise, images were labeled as "normal". Using these materials as training materials for AI deep learning model buildup.	Other: As training material for deep learning model. Using labeled images as training materials for artificial intelligence to develop object detecting model.
rectal cancer lesion images for testing. Using the buildup AI deep learning models from training cohort. Evaluating prediction rate of the model and analysis survival outcomes.	Other: As materials for external validation for the buildup model. Using the external validation set to evaluate prediction rate and survival outcome.

Arm

Intervention/Treatment

rectal cancer lesion images for training

Rectal cancer lesion images. Images with threatened (<2mm) circumferential margin of rectal cancer were labeled as "diseased". Otherwise, images were labeled as "normal". Using these materials as training materials for AI deep learning model buildup.

Other: As training material for deep learning model.

Using labeled images as training materials for artificial intelligence to develop object detecting model.

rectal cancer lesion images for testing.

Using the buildup AI deep learning models from training cohort. Evaluating prediction rate of the model and analysis survival outcomes.

Other: As materials for external validation for the buildup model.

Using the external validation set to evaluate prediction rate and survival outcome.

Outcome Measures

Primary Outcome Measures

accuracy of artificial intelligence with experienced physician [1 week after images done.]
accuracy between artificial intelligence and experienced physician

Secondary Outcome Measures

real life survival outcome of diagnosis by artificial intelligence. [5 years after diagnosed]
real life survival outcome by artificial intelligence.

Eligibility Criteria

Criteria

Ages Eligible for Study:

20 Years to 100 Years

Sexes Eligible for Study:

All

Accepts Healthy Volunteers:

Inclusion Criteria:

clinical staging T3-4 with high quality CT images.

Exclusion Criteria:

1. not primary malignancy lesion
1. not localizing rectum
1. T1-2 lesion
1. non contrast or poor quality images

Contacts and Locations

Locations

	Site	City	State	Country	Postal Code
1	Taichung Verterans General Hospital	Taichung		Taiwan

Sponsors and Collaborators

Taichung Veterans General Hospital

Investigators

Principal Investigator: ChunuYu Lin, M.D., Taichung Veterans General Hospital

Study Documents (Full-Text)

None provided.

More Information

Publications

None provided.

Responsible Party:

Taichung Veterans General Hospital

ClinicalTrials.gov Identifier:

NCT05723965

Other Study ID Numbers:

CE21235B

First Posted:

Feb 13, 2023

Last Update Posted:

Feb 13, 2023

Last Verified:

Feb 1, 2023

Studies a U.S. FDA-regulated Drug Product:

Studies a U.S. FDA-regulated Device Product:

Additional relevant MeSH terms:

Rectal Neoplasms

Study Results

No Results Posted as of Feb 13, 2023