OBJECTIVE: To evaluate a computer-assisted diagnosis system, BONEVAVI version 2 for bone scintigraphy, this study examined the performance of the software in patients with and without skeletal metastasis. METHODS: Bone scans of various patients were analyzed by BONENAVI version 2. Patients with skeletal metastasis from prostate cancer, lung cancer, breast cancer, and other cancers were included in the study as true positive cases. Patients with normal bone scans, consecutive patients with several days of no skeletal metastasis (regardless of hot spots), and patients with abnormal bone scans but no skeletal metastasis were included as negative cases. Patient artificial neural network (ANN) values equal to or above 0.5 were regarded as positive, and those below 0.5 as negative. This study also analyzed cases according to primary cancer factors, osseous metastasis type, and bone tumor burden. RESULTS: The sensitivity of patient ANN values was 121/142 (85 %) for all cancers, 25/29 (86 %) for prostate cancer, 35/40 (88 %) for lung cancer, 37/45 (82 %) for breast cancer, and 24/28 (86 %) for other cancers. The specificity of ANN values was 40/49 (82 %) for normal bone scans, 99/122 (81 %) for consecutive patients with several days of no skeletal metastasis, and 44/81 (54 %) for patients with abnormal bone scans but no skeletal metastasis. Patients showing false negatives included: 10 patients with small lesions (6 of whom showed positive lesion ANN values), 4 patients with osteolytic lesions, 5 patients with intertrabecular osseous metastasis, and 1 patient with a metastatic lesion adjacent to the urinary bladder. The correlation between manually counted lesion numbers and Bone Scan Index was excellent for prostate cancer, and was good for lung cancer, breast cancer, and other cancers. CONCLUSION: BONENAVI version 2 is an effective computer-assisted diagnosis system for bone scanning, but the drawbacks of bone scanning remain unresolved.
OBJECTIVE: To evaluate a computer-assisted diagnosis system, BONEVAVI version 2 for bone scintigraphy, this study examined the performance of the software in patients with and without skeletal metastasis. METHODS: Bone scans of various patients were analyzed by BONENAVI version 2. Patients with skeletal metastasis from prostate cancer, lung cancer, breast cancer, and other cancers were included in the study as true positive cases. Patients with normal bone scans, consecutive patients with several days of no skeletal metastasis (regardless of hot spots), and patients with abnormal bone scans but no skeletal metastasis were included as negative cases. Patient artificial neural network (ANN) values equal to or above 0.5 were regarded as positive, and those below 0.5 as negative. This study also analyzed cases according to primary cancer factors, osseous metastasis type, and bone tumor burden. RESULTS: The sensitivity of patient ANN values was 121/142 (85 %) for all cancers, 25/29 (86 %) for prostate cancer, 35/40 (88 %) for lung cancer, 37/45 (82 %) for breast cancer, and 24/28 (86 %) for other cancers. The specificity of ANN values was 40/49 (82 %) for normal bone scans, 99/122 (81 %) for consecutive patients with several days of no skeletal metastasis, and 44/81 (54 %) for patients with abnormal bone scans but no skeletal metastasis. Patients showing false negatives included: 10 patients with small lesions (6 of whom showed positive lesion ANN values), 4 patients with osteolytic lesions, 5 patients with intertrabecular osseous metastasis, and 1 patient with a metastatic lesion adjacent to the urinary bladder. The correlation between manually counted lesion numbers and Bone Scan Index was excellent for prostate cancer, and was good for lung cancer, breast cancer, and other cancers. CONCLUSION: BONENAVI version 2 is an effective computer-assisted diagnosis system for bone scanning, but the drawbacks of bone scanning remain unresolved.
Authors: Simin Liu; Ming Feng; Tingting Qiao; Haidong Cai; Kele Xu; Xiaqing Yu; Wen Jiang; Zhongwei Lv; Yin Wang; Dan Li Journal: Cancer Manag Res Date: 2022-01-03 Impact factor: 3.989