This international, multicenter retrospective study aims to develop a deep learning (DL)-based predictive model to identify malignant transformation in pancreatic cystic lesions, improving upon current clinical guidelines. The model will integrate clinical, biochemical, and multimodal imaging data. Several 3D convolutional neural networks will be trained using advanced preprocessing, data augmentation, and hybrid fusion techniques. Model performance will be compared to that of existing international guidelines. The study involves no additional procedures for patients and adheres to strict data anonymization and privacy regulations.
Study Type
OBSERVATIONAL
Enrollment
250
Pancreatic resective surgery performed for pancreatic cystic lesions with high risk of malignant degeneration based on clinical, biochemical, and/or radiological features following current guidelines on pancreatic cystic lesions management.
Prediction of malignant degeneration of pancreatic cystics lesions
Predict the presence of malignant degeneration (defined as: high grade dysplasia, in situ PADC, or T1 PADC) in pancreatic cystic lesion(s) using artificial intelligence model based on clinical, biochemical, and radiological features. This will be measured through Area Under the Receiver Operator Characteristic curve (AUROC) assesment. AUROC varies between 0.5 and 1, corresponding to no class separation capacity and full class separation capacity, respectively.
Time frame: 90 days from patients hospital discharge.
Accuracy of performance evaluation
the number of true positives and true negatives among all predictions. It varies between 0 (no correct prediction) to 1 (full correct predictions).
Time frame: 90 days from patients hospital discharge.
Precision of performance evaluation
The number of true positives divided by all the positive predictions (true positives and false positives). It varies between 0 (no correct prediction) to 1 (full correct predictions).
Time frame: 90 days from patients hospital discharge.
Recall of performance evaluation
The number of true positives divided by the actual positive instances in the dataset (true positives and false negatives). It varies between 0 (no correct prediction) to 1 (full correct predictions).
Time frame: 90 days from patients hospital discharge.
Balanced accuracy
the aritmethic mean of sensitivity and specificity. It varies between 0 (no correct prediction) to 1 (full correct predictions).
Time frame: 90 days from patients hospital discharge.
F1-score
It combines precision and recall. It ranges from 0-100%, and a higher F1 score denotes a better quality classifier.
Time frame: 90 days from patients hospital discharge.
Confusion matrix
A visual representation of true positives, false positives, true negatives, and false negatives. It is depicted through a table.
Time frame: 90 days from patients hospital discharge.
Log-loss
It indicates how close the prediction probability is to the corresponding actual/true value (0 or 1 in case of binary classification). The more the predicted probability diverges from the actual value, the higher is the log-loss value.
Time frame: 90 days from patients hospital discharge.
Cohen's Kappa
A metric used to measure the level of agreement between two raters which can be a useful tool to gauge the performance of a classification model. It accounts for the fact that the raters may happen to agree on some items purely by chance. It varies between 0 (no correct prediction) to 1 (full correct predictions).
Time frame: 90 days from patients hospital discharge.
This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.