Real-World Validation of a Computer-Aided Diagnosis System for Prediction of Polyp Histology in Colonoscopy: A Prospective Multicenter Study.

INTRODUCTION: Computer-aided diagnosis (CADx) of polyp histology could support endoscopists in clinical decision-making. However, this has not been validated in a real-world setting. METHODS: We performed a prospective, multicenter study comparing CADx and endoscopist predictions of polyp histology in real-time colonoscopy. Optical diagnosis based on visual inspection of polyps was made by experienced endoscopists. After this, the automated output from the CADx support tool was recorded. All imaged polyps were resected for histological assessment. Primary outcome was difference in diagnostic performance between CADx and endoscopist prediction of polyp histology. Subgroup analysis was performed for polyp size, bowel preparation, difficulty of location of the polyps, and endoscopist experience. RESULTS: A total of 661 eligible polyps were resected in 320 patients aged ≥40 years between March 2021 and July 2022. CADx had an overall accuracy of 71.6% (95% confidence interval [CI] 68.0-75.0), compared with 75.2% (95% CI 71.7-78.4) for endoscopists ( P = 0.023). The sensitivity of CADx for neoplastic polyps was 61.8% (95% CI 56.9-66.5), compared with 70.3% (95% CI 65.7-74.7) for endoscopists ( P < 0.001). The interobserver agreement between CADx and endoscopist predictions of polyp histology was moderate (83.1% agreement, κ 0.661). When there was concordance between CADx and endoscopist predictions, the accuracy increased to 78.1%. DISCUSSION: The overall diagnostic accuracy and sensitivity for neoplastic polyps was higher in experienced endoscopists compared with CADx predictions, with moderate interobserver agreement. Concordance in predictions increased this diagnostic accuracy. Further research is required to improve the performance of CADx and to establish its role in clinical practice.

Gamma knife surgery in the management of brain metastases from lung carcinoma: a retrospective analysis of survival, local tumor control, and freedom from new brain metastasis.

OBJECT: The objective of this retrospective study was to analyze the results of stereotactic radiosurgery performed using a gamma knife in the treatment of 44 consecutive patients with brain metastases from lung carcinoma. METHODS: Forty-four patients with lung carcinoma were treated for metastatic brain tumors by performing radiosurgery with a Leksell Gamma Knife. Twenty-one patients (47.7%) were women and 23 were men. The mean age of the patients was 56 years (range 35-77 years). Twenty-two patients (50%) had solitary tumors and the rest had multiple tumors (two-six lesions). Eighteen patients (40.9%) presented with a recurrent and/or progressive brain disease that previously had been treated with other modalities (surgery, external-beam radiotherapy, or both). Fifteen patients had controlled lung disease and 19 patients had systemic metastases (in lymph nodes, liver, and/or bones) at the time of radiosurgery. The median follow-up period was 18.25 months. All patients were followed up for three different end points: 1) death caused by the disease; 2) clinical and/or radiological evidence of progression of the tumor that had been treated with radiosurgery; and 3) appearance of new lesions. At the last follow-up review, 17 patients (38.6%) were alive and 27 (61.4%) had died. Ten patients (22.7%) died as a result of brain disease (failure of local control or new metastases). Controlled primary disease at the time of detection of metastases and the ability to achieve local tumor control after radiosurgery significantly improved the patient survival (p < 0.01). Control of the treated tumor(s) was achieved in 32 of 44 patients (72 tumors) and 10 patients experienced treatment failure. In addition to the 44 patients comprising the study population, two other patients were treated, but died of lung disease too early in the follow-up period to have been assessed. As of the last follow-up review, no new brain metastasis had occurred in 36 patients (81.8% [includes surviving and nonsurviving patients]). The median duration of overall survival was 7 months, the median period of controlled brain disease was 21 months, and the median period of freedom from new brain metastases was 17 months (95% confidence interval 13-19 months). CONCLUSIONS: Gamma knife surgery has significantly reduced the incidence of mortality from brain disease by effectively accomplishing local tumor control in patients with metastatic lung cancer. Local control and freedom from new brain metastases is not influenced by prior external-beam radiotherapy.