Detection of Local Prostate Cancer Recurrence from PET/CT Scans Using Deep Learning.
Simple Summary: Prostate cancer is a leading cause of cancer-related deaths in men around the world. A type of imaging technique called positron emission tomography (PET), which uses a special scan to detect cancer, has shown great promise in identifying recurring prostate cancer and spread to other...
| Publicado en: | Cancers Vol. 17; no. 9; pp. 1575 - 1613 |
|---|---|
| Autores principales: | , , , , , , , , , , , |
| Formato: | computer program diagnostic images research tables/charts Journal Article |
| Publicado: |
MDPI
May2025
|
| Acceso en línea: | Ver este registro en EBSCOhost |
| fields | @attributes: recordID: 1 pdfLink: plink: https://search.ebscohost.com/login.aspx?direct=true&db=ccm&AN=185133413&site=ehost-live header: @attributes: shortDbName: ccm uiTerm: 185133413 longDbName: CINAHL Complete uiTag: AN controlInfo: bkinfo: dissinfo: jinfo: jid: 20726694 B74B jtl: Cancers issn: 20726694 maglogo: N pubinfo: dt: May2025 vid: 17 iid: 9 pid: 97109 pub: MDPI artinfo: ui: 185133413 185133413 185133413 10.3390/cancers17091575 185133413 ppf: 1575 ppct: 38 formats: tig: atl: Detection of Local Prostate Cancer Recurrence from PET/CT Scans Using Deep Learning. aug: au: Korb, Marko Efetürk, Hülya Jedamzik, Tim Hartrampf, Philipp E. Kosmala, Aleksander Serfling, Sebastian E. Dirk, Robin Michalski, Kerstin Buck, Andreas K. Werner, Rudolf A. Schlötelburg, Wiebke Ankenbrand, Markus J. affil: Center for Computational and Theoretical Biology, Julius-Maximilians-University Würzburg, 97070 Würzburg, Germany sug: subj: Prostatic Neoplasms Diagnosis Neoplasm Recurrence, Local Diagnosis Positron Emission Tomography Computed Tomography Utilization Deep Learning Methods Neural Networks (Computer) Utilization Prostate-Specific Membrane Antigen Program Development Human Funding Source Male Adult Middle Age Aged Aged, 80 and Over Retrospective Design Validation Studies Academic Medical Centers Prostate Anatomy and Histology Convolutional Neural Networks Machine Learning Bladder Prostatectomy Reliability and Validity Descriptive Statistics Data Analysis Software Adult: 19-44 years Middle Aged: 45-64 years Aged: 65+ years Aged, 80 & over Male ab: Simple Summary: Prostate cancer is a leading cause of cancer-related deaths in men around the world. A type of imaging technique called positron emission tomography (PET), which uses a special scan to detect cancer, has shown great promise in identifying recurring prostate cancer and spread to other parts of the body. In this study, we created a computer-based model that uses PET scan images to predict if prostate cancer has come back after treatment. To improve the model's performance, we tried different methods, such as focusing on different parts of the image, adding extra information from the patient's medical history, and including details about whether the patient had prior surgery to remove the prostate. These efforts led to an accuracy of 77% in predicting cancer recurrence. While this accuracy was lower than the desired 90%, the model still showed significant improvement. Many approaches were tested, each helping to improve the model. The results of this study are an important step forward in developing tools that can reliably detect cancer recurrence in the prostate area. However, more research is required to further improve the model's accuracy and make it more useful for doctors in real-life situations. Background: Prostate cancer (PC) is a leading cause of cancer-related deaths in men worldwide. PSMA-directed positron emission tomography (PET) has shown promising results in detecting recurrent PC and metastasis, improving the accuracy of diagnosis and treatment planning. To evaluate an artificial intelligence (AI) model based on [18F]-prostate specific membrane antigen (PSMA)-1007 PET datasets for the detection of local recurrence in patients with prostate cancer. Methods: We retrospectively analyzed 1404 [18F]-PSMA-1007 PET/CTs from patients with histologically confirmed prostate cancer. Artificial neural networks were trained to recognize the presence of local recurrence based on the PET data. First, the hyperparameters were optimized for an initial model (model A). Subsequently, the bladder was localized using an already published model and a model (model B) was trained only on a 20 cm cube around the bladder. Finally, two separate models were trained on the same section depending on the prostatectomy status (model C (post-prostatectomy) and model D (non-operated)). Results: Model A achieved an accuracy of 56% on the validation data. By restricting the region to the area around the bladder, Model B achieved a validation accuracy of 71%. When validating the specialized models according to prostatectomy status, model C achieved an accuracy of 77% and model D an accuracy of 77%. All models achieved accuracies of almost 100% on the training data, indicating overfitting. Conclusions: For the presented task, 1404 examinations were insufficient to reach an accuracy of over 90% even when employing data augmentation, including additional metadata and performing automated hyperparameter optimization. The low F1-score and AUC values indicate that none of the presented models produce reliable results. However, we will facilitate future research and the development of better models by openly sharing our source code and all pre-trained models for transfer learning. pubtype: Academic Journal doctype: computer program diagnostic images research tables/charts Journal Article ougenre: Article language: English refInfo: holdings: @attributes: islocal: N |
|---|