Ignite Creation Date:
2024-07-17 @ 11:24 AM
Last Modification Date:
2024-10-26 @ 3:33 PM
Study NCT ID:
NCT06474338
Status:
RECRUITING
Last Update Posted:
2024-06-25
First Post:
2024-06-19
Brief Title:
AI Detection of Bladder Tumors Under Endoscopy
Sponsor:
Peking Union Medical College Hospital
Organization:
Peking Union Medical College Hospital
Study Overview
Official Title:
Artificial Intelligence Detection of Bladder Tumors Under Endoscopy
Status:
RECRUITING
Status Verified Date:
2024-06
Last Known Status:
None
Delayed Posting:
No
If Stopped, Why?:
Not Stopped
Has Expanded Access:
False
If Expanded Access, NCT#:
N/A
Has Expanded Access, NCT# Status:
N/A
Acronym:
None
Brief Summary:
The goal of this clinical trial is to learn if the AI algorithm can detect bladder tumors better than urologists under cystoscopy It will also train the AI algorithm for bladder tumor detection The main question it aims to answer is
Can AI algorithm achieve IOU value precision recall false negative rate of bladder tumor detection similar to that of urologists The cystoscopy video will be annotated by AI and urologists Researchers will compare AI algorithm to urologists to see if Al algorithm has a similar capability as urologists do
Can AI algorithm achieve IOU value precision recall false negative rate of bladder tumor detection similar to that of urologists The cystoscopy video will be annotated by AI and urologists Researchers will compare AI algorithm to urologists to see if Al algorithm has a similar capability as urologists do
Detailed Description:
1 Research Background Bladder cancer is the ninth most common malignancy worldwide with an estimated 430000 new cases diagnosed annually The standard diagnosis and monitoring of bladder cancer rely on white light cystoscopy WLC with over 2 million cystoscopies performed annually in the United States and Europe Due to the high recurrence rate of bladder cancer frequent monitoring and intervention are necessary
Early detection and complete resection of non-muscle invasive bladder cancer can reduce recurrence and progression However up to 40 of patients with multifocal disease do not achieve complete resection during the initial transurethral resection of bladder tumor TURBT Many papillary tumors and flat lesions are difficult to identify through WLC There is an urgent need for cost-effective non-invasive and user-friendly adjunct imaging technologies to address the diagnostic deficiencies of WLC
Recent advancements in deep learning-based automated image processing may provide new solutions to the limitations of cystoscopy Convolutional neural networks CNNs possess the ability to learn complex relationships and integrate existing knowledge into models showing potential applications across various fields including bladder tumor diagnosis We employed the HRNet algorithm a convolutional neural network for enhanced bladder tumor detection
2 Research Objectives We aim to explore the potential application of AI in urological tumors by collecting cystoscopy videos from patients undergoing cystoscopy These videos include bladder tumors will be annotated manually by urologists then the AI algorithm will be used to recognize the bladder tumors
3 Research Methods This is a multicenter retrospective observational study
4 Research Process 41 Patient Cohort Inclusion criteria 1 The patients who had bladder tumor and received WLC or TURBT and the full-length surgery video is available
Exclusion criteria 1 The video is too blurry to distinguish the normal bladder wall and bladder tumor 2 Lack of the appearance of bladder tumor before resection 3 Lack of informed consent
Patient information in the videos will not be shown Videos from the initially recruited 200 bladder tumor patients will be used for algorithm development Videos from an additional 100 patients are used for algorithm validation
42 Data Preprocessing To reduce the data volume we extract the frame at a ratio of 14 Two urologists outline the boundary of bladder tumors in each frame seperately and check for each other AI algorithm is used to contour the same bladder tumors The outlines of bladder tumor annotated by urologists and algorithm are compared the IOU value precision sensitivity and false negative rate are analyzed
43 Algorithm Development This study uses semantic segmentation to identify bladder tumors in WLC The D-LinkNet network structure used a pre-trained ResNet34 on the ImageNet dataset as its encoder with the central part utilizing dilated convolutions with different dilation rates in a cascaded manner and upsampling performed using deconvolution The original resolution of all images are 19201080 downsampled by 2 to 960540 and zero-padded in the width direction to obtain the image with a resolution of 960544 The RGB images of this size were normalized mean-subtracted and variance-divided before being input into the network The images undergo five encoding processes dilated convolutions and five decoding processes ultimately producing a prediction result of 960544 which was further post-processed In this research the parameter settings were as follows batch size of 8 Adam optimizer a learning rate of 0001 The training environment was an NVIDIA TITAN Xp GPU
44 Results Interpretation The intersection over union IOU is a crucial standard for evaluating single-frame image recognition capability in image recognition When IOU is above the threshold it suggests that the model detects the object successfully indicating a true positive When IOU can not reach the threshold it suggests that the model fails to detect the object and indicating a false negative If a prediction appears without ground truth in the image it is considered a false positive We calculate the models sensitivity and precision in the test set The Dice coefficient measures the similarity between two samples A higher average Dice coefficient indicates a better detection performance of the model
45 Observation Indicators
① Video annotation and classification annotation status and RLN recognition discernibility classification results for training and test set surgery videos ② After grouping and training observe the models sensitivity precision false negative rate false positive rate and average Dice coefficient at IoU thresholds of 01 and 05 in the test set under different discernibility groups
46 Statistical Methods Analysis will be analyzed by R 402 software the data will be expressed as absolute numbers or percentages
5 Data Management and Confidentiality All data in this study is properly stored to ensure security without loss or leakage Sensitive information and patient information will not be uploaded to public platforms During data processing patients personal information will be anonymized and patient identification codes will be used to replace patient names and IDs If technical services are needed data will be appropriately encrypted and a confidentiality agreement will be signed
Early detection and complete resection of non-muscle invasive bladder cancer can reduce recurrence and progression However up to 40 of patients with multifocal disease do not achieve complete resection during the initial transurethral resection of bladder tumor TURBT Many papillary tumors and flat lesions are difficult to identify through WLC There is an urgent need for cost-effective non-invasive and user-friendly adjunct imaging technologies to address the diagnostic deficiencies of WLC
Recent advancements in deep learning-based automated image processing may provide new solutions to the limitations of cystoscopy Convolutional neural networks CNNs possess the ability to learn complex relationships and integrate existing knowledge into models showing potential applications across various fields including bladder tumor diagnosis We employed the HRNet algorithm a convolutional neural network for enhanced bladder tumor detection
2 Research Objectives We aim to explore the potential application of AI in urological tumors by collecting cystoscopy videos from patients undergoing cystoscopy These videos include bladder tumors will be annotated manually by urologists then the AI algorithm will be used to recognize the bladder tumors
3 Research Methods This is a multicenter retrospective observational study
4 Research Process 41 Patient Cohort Inclusion criteria 1 The patients who had bladder tumor and received WLC or TURBT and the full-length surgery video is available
Exclusion criteria 1 The video is too blurry to distinguish the normal bladder wall and bladder tumor 2 Lack of the appearance of bladder tumor before resection 3 Lack of informed consent
Patient information in the videos will not be shown Videos from the initially recruited 200 bladder tumor patients will be used for algorithm development Videos from an additional 100 patients are used for algorithm validation
42 Data Preprocessing To reduce the data volume we extract the frame at a ratio of 14 Two urologists outline the boundary of bladder tumors in each frame seperately and check for each other AI algorithm is used to contour the same bladder tumors The outlines of bladder tumor annotated by urologists and algorithm are compared the IOU value precision sensitivity and false negative rate are analyzed
43 Algorithm Development This study uses semantic segmentation to identify bladder tumors in WLC The D-LinkNet network structure used a pre-trained ResNet34 on the ImageNet dataset as its encoder with the central part utilizing dilated convolutions with different dilation rates in a cascaded manner and upsampling performed using deconvolution The original resolution of all images are 19201080 downsampled by 2 to 960540 and zero-padded in the width direction to obtain the image with a resolution of 960544 The RGB images of this size were normalized mean-subtracted and variance-divided before being input into the network The images undergo five encoding processes dilated convolutions and five decoding processes ultimately producing a prediction result of 960544 which was further post-processed In this research the parameter settings were as follows batch size of 8 Adam optimizer a learning rate of 0001 The training environment was an NVIDIA TITAN Xp GPU
44 Results Interpretation The intersection over union IOU is a crucial standard for evaluating single-frame image recognition capability in image recognition When IOU is above the threshold it suggests that the model detects the object successfully indicating a true positive When IOU can not reach the threshold it suggests that the model fails to detect the object and indicating a false negative If a prediction appears without ground truth in the image it is considered a false positive We calculate the models sensitivity and precision in the test set The Dice coefficient measures the similarity between two samples A higher average Dice coefficient indicates a better detection performance of the model
45 Observation Indicators
① Video annotation and classification annotation status and RLN recognition discernibility classification results for training and test set surgery videos ② After grouping and training observe the models sensitivity precision false negative rate false positive rate and average Dice coefficient at IoU thresholds of 01 and 05 in the test set under different discernibility groups
46 Statistical Methods Analysis will be analyzed by R 402 software the data will be expressed as absolute numbers or percentages
5 Data Management and Confidentiality All data in this study is properly stored to ensure security without loss or leakage Sensitive information and patient information will not be uploaded to public platforms During data processing patients personal information will be anonymized and patient identification codes will be used to replace patient names and IDs If technical services are needed data will be appropriately encrypted and a confidentiality agreement will be signed
Study Oversight
Has Oversight DMC:
None
Is a FDA Regulated Drug?:
False
Is a FDA Regulated Device?:
False
Is an Unapproved Device?:
None
Is a PPSD?:
None
Is a US Export?:
None
Is an FDA AA801 Violation?:
None