Background: Pancreatic cysts in autosomal dominant polycystic kidney disease (ADPKD) correlate with PKD2 mutations, which have a different phenotype than PKD1 mutations. However, pancreatic cysts are commonly overlooked by radiologists. Here, we automate the detection of pancreatic cysts on abdominal MRI in ADPKD.
Methods: Eight nnU-Net-based segmentation models with 2D or 3D configuration and various loss functions were trained on positive-only or positive-and-negative datasets, comprising axial and coronal T2-weighted MR images from 254 scans on 146 ADPKD patients with pancreatic cysts labeled independently by two radiologists. Model performance was evaluated on test subjects unseen in training, comprising 40 internal, 40 external, and 23 test-retest reproducibility ADPKD patients.
Results: Two radiologists agreed on 52% of cysts labeled on training data, and 33%/25% on internal/external test datasets. The 2D model with a loss of combined dice similarity coefficient and cross-entropy trained with the dataset with both positive and negative cases produced an optimal dice score of 0.7 ± 0.5/0.8 ± 0.4 at the voxel level on internal/external validation and was thus used as the best-performing model. In the test-retest, the optimal model showed superior reproducibility (83% agreement between scan A and B) in segmenting pancreatic cysts compared to six expert observers (77% agreement). In the internal/external validation, the optimal model showed high specificity of 94%/100% but limited sensitivity of 20%/24%.
Conclusions: Labeling pancreatic cysts on T2 images of the abdomen in patients with ADPKD is challenging, deep learning can help the automated detection of pancreatic cysts, and further image quality improvement is warranted.
Download full-text PDF |
Source |
---|---|
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC11281294 | PMC |
http://dx.doi.org/10.3390/tomography10070087 | DOI Listing |
J Cyst Fibros
January 2025
Leeds Institute of Medical Research, University of Leeds, School of Medicine, Leeds, United Kingdom; The Leeds Adult CF Unit, Leeds Teaching Hospitals NHS Trust, Leeds, United Kingdom. Electronic address:
Background: Whether improvements in gastrointestinal (GI) symptoms observed with Elexacaftor/Tezacaftor/Ivacaftor (ETI) treatment are sustained in the longer-term requires exploration. This study investigated how GI-symptoms change with longer-term ETI use in pancreatic insufficient adults with cystic fibrosis (awCF).
Methods: Participants completed up to three abdominal symptom questionnaires, employing the validated CFAbd-Score.
Intern Med
January 2025
Department of Nephrology, The University of Osaka Graduate School of Medicine, Japan.
We encountered a family with hereditary renal failure, renal medullary cysts, pancreatic hypoplasia, hypomagnesemia, liver enzyme abnormalities, and diabetes mellitus (DM). We identified a novel heterozygous variant of HNF1B (NM_000458.4:c.
View Article and Find Full Text PDFHealth Sci Rep
January 2025
Gerhard-Domagk Institute of Pathology University Hospital Muenster (UKM) Muenster Germany.
Background And Aims: Benign lesions, inflammation, cysts and pseudocysts, as well as neoplasms of the exocrine and endocrine parts of the pancreas can be easily identified using cytological methods. The sensitivity and specificity can be increased with the help of additional examination methods. The sensitivity of intraoperative rapid cytology reaches about 99%.
View Article and Find Full Text PDFEndosc Ultrasound
December 2024
Department of Gastroenterology, the First Medical Center, Chinese PLA General Hospital, Beijing 100083, China.
Background And Objectives: An accurate diagnosis is crucial for the clinical management of pancreatic cystic neoplasm (PCN). EUS-guided through-the-needle biopsy (EUS-TTNB) is a novel technique for improving the accuracy of PCN diagnosis. There is insufficient evidence about the efficacy of EUS-TTNB.
View Article and Find Full Text PDFJ Gastrointest Surg
January 2025
Department of General Surgery, Cleveland Clinic Foundation, Cleveland, OH.
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!