Search

Rising incidence and mortality of cancer have led to an incremental amount of research in the field. To learn from preexisting data, it has become important to capture maximum information related to disease type, stage, treatment, and outcomes. Medical imaging reports are rich in this kind of information but are only present as free text. The extraction of information from such unstructured text reports is labor-intensive. The use of Natural Language Processing (NLP) tools to extract information from radiology reports can make it less time-consuming as well as more effective. In this study, we have developed and compared different models for the classification of lung carcinoma reports using clinical concepts. This study was approved by the institutional ethics committee as a retrospective study with a waiver of informed consent. A clinical concept-based classification pipeline for lung carcinoma radiology reports was developed using rule-based as well as machine learning models and compared. The machine learning models used were XGBoost and two more deep learning model architectures with bidirectional long short-term neural networks. A corpus consisting of 1700 radiology reports including computed tomography (CT) and positron emission tomography/computed tomography (PET/CT) reports were used for development and testing. Five hundred one radiology reports from MIMIC-III Clinical Database version 1.4 was used for external validation. The pipeline achieved an overall F1 score of 0.94 on the internal set and 0.74 on external validation with the rule-based algorithm using expert input giving the best performance. Among the machine learning models, the Bi-LSTM_dropout model performed better than the ML model using XGBoost and the Bi-LSTM_simple model on internal set, whereas on external validation, the Bi-LSTM_simple model performed relatively better than other 2. This pipeline can be used for clinical concept-based classification of radiology reports related to lung carcinoma from a huge corpus and also for automated annotation of these reports., (© 2023. The Author(s).)

Purpose: To analyze diagnostic performance of F-18 FDG PET/CT in recurrent adenocarcinoma gallbladder (GBC) and to establish its possible impact on post-recurrence survival., Method: FDG PET/CT studies of suspected recurrent GBC were retrospectively analyzed alongside tumor markers serum CEA and CA 19-9. Abnormal FDG-avid lesions and abnormal morphological lesions were considered positive for recurrence, and were categorized as isolated abdominal wall recurrence, loco-regional recurrence, and distant metastatic disease. Histopathology, definite progression on imaging and positive response to treatment was considered as reference standard. Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and accuracy were used as diagnostic performance parameters. Post-recurrence survival was calculated whenever appropriate follow-up was available, based on the abovementioned categories of sites of recurrence using survival curves and log-rank test., Results: Out of 117 PET/CT studies, 93 (79.5%) were positive and 24 (20.5%) were negative for recurrence. 86 out of 93 were true positive and 23 of 24 were true negative. PET/CT demonstrated sensitivity, specificity, PPV, NPV and accuracy of 98.8%, 76.7%, 92.5%, 95.8% and 93.1%, respectively. Diagnostic performance of PET/CT was significantly better than combination tumor markers. Of 66 cases with available follow-up, isolated abdominal wall (port/scar site) recurrence and loco-regional recurrence demonstrated significantly higher post-recurrence survival as compared to distant metastasis; median survival being 39, 25 and 12 months, respectively., Conclusion: F-18 FDG PET/CT has better diagnostic performance than tumor markers combination. Isolated abdominal wall (port/scar site) recurrence and loco-regional recurrence on PET/CT demonstrated better survival than non-regional metastatic disease. These results suggest a possible role of PET/CT as a surveillance modality, as well as a guide to therapeutic decision-making in cases of recurrent GBC., (© 2022. The Author(s) under exclusive licence to Japan Radiological Society.)

Recognition of patterns has always been extremely important in cross-sectional imaging. Peritoneal involvement, both primary and as dissemination from abdominopelvic malignancies, is manifested in different forms, purely because of anatomical complexity. We studied series of peritoneal involvement by (18)F-fluorodeoxyglucose positron-emission tomography/computed tomography fusion imaging and derived patterns of tracer uptake on maximum intensity projection and cross-sectional fusion images.