Dataset Information

Deep learning-based classification of blue light cystoscopy imaging during transurethral resection of bladder tumors.

ABSTRACT: Bladder cancer is one of the top 10 frequently occurring cancers and leads to most cancer deaths worldwide. Recently, blue light (BL) cystoscopy-based photodynamic diagnosis was introduced as a unique technology to enhance the detection of bladder cancer, particularly for the detection of flat and small lesions. Here, we aim to demonstrate a BL image-based artificial intelligence (AI) diagnostic platform using 216 BL images, that were acquired in four different urological departments and pathologically identified with respect to cancer malignancy, invasiveness, and grading. Thereafter, four pre-trained convolution neural networks were utilized to predict image malignancy, invasiveness, and grading. The results indicated that the classification sensitivity and specificity of malignant lesions are 95.77% and 87.84%, while the mean sensitivity and mean specificity of tumor invasiveness are 88% and 96.56%, respectively. This small multicenter clinical study clearly shows the potential of AI based classification of BL images allowing for better treatment decisions and potentially higher detection rates.

SUBMITTER: Ali N

PROVIDER: S-EPMC8172542 | biostudies-literature | 2021 Jun

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Deep learning-based classification of blue light cystoscopy imaging during transurethral resection of bladder tumors.

Ali Nairveen N Bolenz Christian C Todenhöfer Tilman T Stenzel Arnulf A Deetmar Peer P Kriegmair Martin M Knoll Thomas T Porubsky Stefan S Hartmann Arndt A Popp Jürgen J Kriegmair Maximilian C MC Bocklitz Thomas T

Scientific reports 20210602 1

Bladder cancer is one of the top 10 frequently occurring cancers and leads to most cancer deaths worldwide. Recently, blue light (BL) cystoscopy-based photodynamic diagnosis was introduced as a unique technology to enhance the detection of bladder cancer, particularly for the detection of flat and small lesions. Here, we aim to demonstrate a BL image-based artificial intelligence (AI) diagnostic platform using 216 BL images, that were acquired in four different urological departments and patholo ...[more]

PMID: 34079004

Similar Datasets

Project description:BackgroundTransurethral resection of bladder tumor (TURBT) is a central component in the diagnosis of non-muscle-invasive bladder cancer (NMIBC) and can be guided by several optical imaging techniques for better visualization of lesions.ObjectiveTo investigate if a change in tumor microenvironment (TME) composition could be observed as an effect of hexaminolevulinate (HAL)-assisted blue light cystoscopy (BLC) in TURBT samples from patients with bladder cancer.Design setting and participantsThis was a retrospective study of 40 patients with bladder cancer who underwent either BLC-guided TURBT (n = 20) or white light cystoscopy (WLC)-guided TURBT (n = 20) before radical cystectomy (RC). Tissue samples (n = 80) were collected from paired TURBT and RC specimens for all 40 patients. Tumor tissue was stained using multiplex immunofluorescence and immunohistochemistry.Outcome measurements and statistical analysisImmune cell infiltration was assessed according to the proportions of each immune cell or immune evasion marker and the relative change from TURBT as baseline was calculated. Statistical comparisons between groups were performed using the Wilcoxon rank-sum test or the paired-sample Wilcoxon test.Results and limitationsComparison of relative changes in the TME revealed a significant decrease in stromal infiltration of cytotoxic T cells (p = 0.024), B cells (p = 0.041), and stromal cells expressing PD-1 (p = 0.011) in patients treated with BLC-guided TURBT compared to WLC-guided TURBT.ConclusionsOur pilot study showed that HAL-BLC during TURBT in bladder cancer may influence the immune cell composition and TME.Patient summaryWe investigated the potential therapeutic effect of blue light versus white light for guidance in removing bladder tumors via the urethra in patients with bladder cancer. For blue light guidance, a compound called hexaminolevulinate is used to visualize tumor tissue. We found changes in immune cell composition that may have been influenced by the blue light guidance.

Project description:BackgroundTransurethral resection of bladder tumor (TURBT) is an essential procedure both for the treatment and staging of bladder cancer, particularly non-muscle invasive bladder cancer (NMIBC). The dissemination of cancer cells during resection and the consequent seeding into the bladder mucosa is the main cause of post-TURBT intravesical recurrence. Although the tumor dissemination is inevitable during conventional TURBT (cTURBT), this drawback can be overcome by tumor resection in one piece with intact surrounding normal tissues, referred to as en bloc resection. We previously described the photodynamic diagnosis (PDD)-assisted en bloc TURBT (EBTUR) technique and its favorable outcomes. Based on our preliminary studies, this randomized controlled trial was designed to evaluate the superiority of PDD-EBTUR to PDD-cTURBT.MethodsThe FLEBER study is a single-center randomized controlled trial in NMIBC patients who require TURBT. The longest diameter of the tumor must be between 6 and 30 mm. A total of 160 eligible patients will be enrolled after screening and randomly allocated to the PDD-EBTUR (experimental) and PDD-cTURBT (control) groups in a 1:1 ratio (80 cases to 80 cases). All patients will be treated using a single, immediate postoperative intravesical chemotherapy with epirubicin. The primary endpoint of this trial is the 2-year recurrence-free survival after surgery in pathologically proven low- or intermediate-risk NMIBC. All patients will be monitored by cystoscopy and urine cytology every 3 months for 2 years. Patient data including adverse events and complications, and data from frequency volume charts, pain scales, and health-related QOL questionnaires will be collected before and after the TURBT at indicated visits.DiscussionThe goal of this trial is to determine the potential benefits of PDD-cTURBT and PDD-EBTUR followed by a single immediate postoperative intravesical chemotherapy in patients with low- or intermediate-risk NMIBC who undergo TURBT. Ultimately, our findings will lead to the development of better interventions and potentially change the standard of care.Trial registrationThis clinical trial was prospectively registered with the UMIN Clinical Trials Registry on 1 August 2020. The reference number is UMIN000041273 , and the Ethics Committee of Nara Medical University Approval ID is 2702.

Dataset Information

Deep learning-based classification of blue light cystoscopy imaging during transurethral resection of bladder tumors.

Publications

Deep learning-based classification of blue light cystoscopy imaging during transurethral resection of bladder tumors.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets