Enhanced diagnostic accuracy for quantitative bone scan using an artificial neural network system: a Japanese multi-center database project.

BACKGROUND: Artificial neural network (ANN)-based bone scan index (BSI), a marker of the amount of bone metastasis, has been shown to enhance diagnostic accuracy and reproducibility but is potentially affected by training databases. The aims of this study were to revise the software using a large number of Japanese databases and to validate its diagnostic accuracy compared with the original Swedish training database. METHODS: The BSI was calculated with EXINIbone (EB; EXINI Diagnostics) using the Swedish training database (n = 789). The software using Japanese training databases from a single institution (BONENAVI version 1, BN1, n = 904) and the revised version from nine institutions (version 2, BN2, n = 1,532) were compared. The diagnostic accuracy was validated with another 503 multi-center bone scans including patients with prostate (n = 207), breast (n = 166), and other cancer types. The ANN value (probability of abnormality) and BSI were calculated. Receiver operating characteristic (ROC) and net reclassification improvement (NRI) analyses were performed. RESULTS: The ROC analysis based on the ANN value showed significant improvement from EB to BN1 and BN2. In men (n = 296), the area under the curve (AUC) was 0.877 for EB, 0.912 for BN1 (p = not significant (ns) vs. EB) and 0.934 for BN2 (p = 0.007 vs. EB). In women (n = 207), the AUC was 0.831 for EB, 0.910 for BN1 (p = 0.016 vs. EB), and 0.932 for BN2 (p < 0.0001 vs. EB). The optimum sensitivity and specificity based on BN2 was 90% and 84% for men and 93% and 85% for women. In patients with prostate cancer, the AUC was equally high with EB, BN1, and BN2 (0.939, 0.949, and 0.957, p = ns). In patients with breast cancer, the AUC was improved from EB (0.847) to BN1 (0.910, p = ns) and BN2 (0.924, p = 0.039). The NRI using ANN between EB and BN1 was 17.7% (p = 0.0042), and that between EB and BN2 was 29.6% (p < 0.0001). With respect to BSI, the NRI analysis showed downward reclassification with total NRI of 31.9% ( p < 0.0001). CONCLUSION: In the software for calculating BSI, the multi-institutional database significantly improved identification of bone metastasis compared with the original database, indicating the importance of a sufficient number of training databases including various types of cancers.

Deep-inspiration breath-hold PET/CT versus free breathing PET/CT and respiratory gating PET for reference: evaluation in 95 patients with lung cancer.

OBJECTIVE: The objective of this study was to define the factors that correlate with differences in maximum standardized uptake value (SUV(max)) in deep-inspiration breath-hold (DIBH) and free breathing (FB) PET/CT admixed with respiratory gating (RG) PET for reference. METHODS: Patients (n = 95) with pulmonary lesions were evaluated at one facility over 33 months. After undergoing whole-body PET/CT, a RG PET and FB PET/CT scans were obtained, followed by a DIBH PET/CT scan. All scans were recorded using a list-mode dynamic collection method with respiratory gating. The RG PET was reconstructed using phase gating without attenuation correction; the FB PET was reconstructed from the RG PET sinogram datasets with attenuation correction. Respiratory motion distance, breathing cycle speed, and waveform of RG PET were recorded. The SUV(max) of FB PET/CT and DIBH PET/CT were recorded: the percent difference in SUV(max) between the FB and DIBH scans was defined as the %BH-index. RESULTS: The %BH-index was significantly higher for lesions in the lower lung area than in the upper lung area. Respiratory motion distance was significantly higher in the lower lung area than in the upper lung area. A significant relationship was observed between the %BH-index and respiratory motion distance. Waveforms without steady end-expiration tended to show a high %BH-index. Significant inverse relationships were observed between %BH-index and cycle speed, and between respiratory motion distance and cycle speed. CONCLUSION: Decrease in SUV(max) of FB PET/CT was due to (1) tumor size, (2) distribution of lower lung, (3) long respiratory movement at slow breathing cycle speeds, and (4) respiratory waveforms without steady end-expiration.

Deep-inspiration breath-hold PET/CT of lung cancer: maximum standardized uptake value analysis of 108 patients.

UNLABELLED: Our aim was to compare the maximum standardized uptake value (SUVmax) between breath-hold (BH) PET/CT and free-breathing (FB) PET/CT. METHODS: The features of phantom data were analyzed, after which a clinical study was performed. A total of 108 consecutive patients with lung cancer were examined using lutetium oxyorthosilicate (LSO)-based PET/CT. The patients were instructed to breathe freely during FB PET/CT. In BH PET/CT, the patients were instructed to hold their breath in the maximal inspiration position during the scout scan, for 10 s of the CT scan, and for as long as possible during the PET scan. BH time was recorded using a respiratory monitoring device. The %BH-index was defined as the percentage difference between SUVmax of FB PET and that of BH PET. Statistical analyses were performed using the following factors: %BH-index, age, body mass index, 18F-FDG dosage, blood glucose, BH time, lesion size, and location. RESULTS: The highest %BH-index was 223.2. %BH-index in the lower lung area was significantly higher than that in the upper lung area (51.8 +/- 49.5 vs. 16.9 +/- 25.6, respectively). Lesion volume and maximum diameter in the high-%BH-index group were significantly lower than those in the low-%BH-index group, with the use of a %BH-index cutoff value of 37.l. CONCLUSION: SUVmax of FB PET should not be taken as accurate, especially in the lower lung area and for small pulmonary lesions. BH PET/CT is expected to enable precise measurement of SUVmax and is thus recommended as part of the standard protocol for lung cancer.

Sentinel lymphoscintigraphy in patients with breast cancer undergoing excisional biopsy.

UNLABELLED: Some small tumors of the breast cannot be diagnosed by needle biopsy, and an excisional biopsy is occasionally needed for the diagnosis. Sentinel node navigation surgery is frequently suitable for patients with such small breast cancers. The purpose of this study was to compare sentinel lymphoscintigrams in breast cancer patients who had previously undergone excisional biopsy with sentinel lymphoscintigrams in patients undergoing no excisional biopsy. We also investigated the possibility of clinical application of the sentinel node navigation procedure in the former group of patients. METHODS: Sentinel lymphoscintigrams of 43 patients with breast cancer undergoing excisional biopsy were compared to those of 116 patients without excisional biopsy. Lymphoscintigrams were obtained by using intradermal and/or subdermal injections of technetium-99m labeled phytate at 2 points on each side of the dermal incision in patients after excisional biopsy. Injections were performed at 2 points of the skin over the tumor in the patients who had not undergone excisional biopsy. RESULTS: Axillary lymph nodes were visualized in 42 of 43 patients undergoing excisional biopsy (98%) and in 115 of 116 patients without excisional biopsy (99%). The number of visualized axillary nodes was 1 to 5 (mean +/- SD = 2.1 +/- 1.0) and 1 to 5 (mean +/- SD = 1.9 +/- 1.0) in the two groups, respectively. No significant difference was determined between the two groups. Parasternal lymph nodes were depicted in 3 patients after excisional biopsy who had the tumor in the outer half of the breast, in contrast to 4 without excisional biopsy who had the tumor in the inner half. Intramammary hot spots were observed in 5 patients after excisional biopsy and in 2 without excisional biopsy. Lymphatic vessels were observed in 23 patients (53%) who had the excisional biopsy, and in 37 (32%) who did not have the biopsy. The former figure was significantly higher than the latter (p < 0.02). CONCLUSION: Sentinel node navigation surgery for axillary nodes was shown to be possible in patients undergoing excisional biopsy. However, the visualization of parasternal nodes, intramammary hot spots and lymphatic vessels tended to increase in number, and care must be exercised in the management of these patients.