Unique ID issued by UMIN | UMIN000042452 |
---|---|
Receipt number | R000048461 |
Scientific Title | Evaluation of postoperative hemorrhage in cardiac surgery using image analysis system and Machine learning |
Date of disclosure of the study information | 2020/12/01 |
Last modified on | 2021/05/05 12:00:22 |
Evaluation of postoperative hemorrhage in cardiac surgery using image analysis system and Machine learning
Evaluation of postoperative hemorrhage in cardiac surgery using image analysis system and Machine learning
Evaluation of postoperative hemorrhage in cardiac surgery using image analysis system and Machine learning
Evaluation of postoperative hemorrhage in cardiac surgery using image analysis system and Machine learning
Japan |
Patients who have undergone open heart surgery
Cardiovascular surgery |
Others
NO
The purpose of this project is to create a classifier (system) that can discriminate post-operative blood loss and cases requiring reopening of the chest in open heart surgery from surgical images using an image analysis system and machine learning.
Efficacy
To create a classifier (system) that can identify cases that require reopening of the heart in open heart surgery. Then, create a receiver operating characteristic curve for the classifier and determine the area under the curve, sensitivity and specificity.
Observational
20 | years-old | <= |
Not applicable |
Male and Female
Patients who underwent open heart surgery at Yamagata University Hospital between January 2010 and December 2020 will be included
Cases with missing image data
Cases with missing blood loss data
500
1st name | Tatsuya |
Middle name | |
Last name | Hayasaka |
Yamagata University Faculty of Medicine
Department of Anesthesia
9909585
2-2-2 Iida-nishi, Yamagata city, Yamagata 990-9585, Japan
023-628-5400
hayasakatatsuya1101@gmail.com
1st name | tatsuya |
Middle name | |
Last name | hayasaka |
Yamagata University Faculty of Medicine
Department of Anesthesia
9909585
2-2-2 Iida-nishi, Yamagata city, Yamagata 990-9585, Japan
023-628-5400
hayasakatatsuya1101@gmail.com
Yamagata University Faculty of Medicine
Department of Anesthesiology, Yamagata University Faculty of Medicine
Self funding
The Ethical Review Committee of Yamagata University Faculty of Medicine
2-2-2 Iida-nishi, Yamagata city, Yamagata 990-9585, Japan
0236285015
ikekenkyu@jm.kj.yamagata-u.ac.jp
NO
2020 | Year | 12 | Month | 01 | Day |
Unpublished
Open public recruiting
2020 | Year | 11 | Month | 01 | Day |
2020 | Year | 10 | Month | 31 | Day |
2020 | Year | 12 | Month | 01 | Day |
2030 | Year | 12 | Month | 31 | Day |
Patients who had open heart surgery performed at Yamagata University Hospital between January 2010 and December 2020 will be included in the study. Videos of open heart surgery performed during the above period are reviewed, and the images just before chest closure are extracted. The images are subjected to machine learning using Scikit-Learn to create a system for supervised learning that predicts the blood loss in the 12 hours after the surgery from the images immediately before chest closure. We use images as input and bleeding volume as output, with 80% of the data as train data and 20% as test data.
As a secondary evaluation, we will create a system for predicting postoperative blood loss using preoperative patient data (age, sex, height, weight, temperature, heart rate, blood pressure, oxygenation capacity, electrolytes, renal function, liver function, white blood cell count, red blood cell count, platelet count, fibrinogen concentration, cardiac function, etc.).
2020 | Year | 11 | Month | 13 | Day |
2021 | Year | 05 | Month | 05 | Day |
Value
https://center6.umin.ac.jp/cgi-open-bin/ctr_e/ctr_view.cgi?recptno=R000048461
Research Plan | |
---|---|
Registered date | File name |
Research case data specifications | |
---|---|
Registered date | File name |
Research case data | |
---|---|
Registered date | File name |