Prospective evaluation of computer-assisted analysis of skeletal lesions for the staging of prostate cancer

Abstract

BackgroundThe purpose of this study was to compare the agreement of the bone scan index (BSI) using EXINI BoneBSI versus experts’ readings in the initial staging for bone metastasis in prostate cancer. In addition, the diagnostic outcome was assessed in a large subset of patients where a true reference for metastases could be determined based on clinical and biochemical follow-up and/or supplementary imaging.MethodsA total of 342 patients had a bone scintigraphy as part of routine staging for prostate cancer. Supplementary imaging was obtained at the discretion of the referring urologist. After full recruitment, the BSI and the number of malignant lesions were calculated using EXINI BoneBSI, and three imaging experts independently classified bone status by a dichotomous outcome (M1 for bone metastasis, M0 for no bone metastasis). A true reference was available in a subset of the patients based on post-operative prostate-specific antigen responses after radical prostatectomy and/or supplementary imaging.ResultsSoftware analysis with a BSI > 0 as the cut-off for metastasis showed excellent agreement with expert classification for M1 disease (96% of the patients) but modest agreement for M0 disease (38%). With a BSI > 1, the agreement was 58% for M1 and 98% for M0. Software analyses based on individual European Association of Urology risk classification did not improve the diagnostic performance. Among patients with a true reference, the software showed metastasis in 64% of the M0 patients but correctly classified metastases in all M1 patients. The sensitivity was 100%, the specificity was 36%, the positive predictive value was 12.6% and the negative predictive value was 100% with a BSI >0 compared with 66.7%, 97.8%, 72.7%, and 97.0% with a BSI > 1.ConclusionThe diagnostic value of using EXINI Bone for the BSI in the staging of newly diagnosed prostate cancer is limited.

Analysis of Bone Scans in Various Tumor Entities Using a Deep-Learning-Based Artificial Neural Network Algorithm—Evaluation of Diagnostic Performance

“…The specificity in prostate cancer increased (68 to 98.6%) at the cost of a moderate reduction in sensitivity (92 to 87%). Using optimized BSI cut-off values, we observed increased performance in other tumor entities (CRC, HCC, and UCC) up to a certain extent, comparable to results for prostate cancer patients [19,34]. The positive predictive value was partially poor (breast cancer 52%, lung cancer 26%, and HCC 53%), demonstrating that the method is still not readily applicable for the individual confirmation of metastatic disease in these entities.…”

“…Many scans therefore had a positive definite BSI due to a solitary area of increased uptake, even though there was no obviously suspicious uptake according to the physician’s statement. This effect was already analyzed by Petersen et al [ 34 ], who used BSI cut-off values of 0% and 1% to improve automated M0/M1 differentiation in prostate cancer patients and compare the results to expert opinion. Using a BSI cut-off of 0%, sensitivity (96%) and mean BSI in M0/M1 cohorts were comparable to our results, but specificity (38%) and PPV (21%) were lower.…”

Evaluation of changes in Bone Scan Index at different acquisition time‐points in bone scintigraphy

“…We have not found any studies investigating if BSI is affected using different radiopharmaceuticals, but published studies on BSI have used different traders, such as HDP (Poulsen et al ., ; Lindgren Belal et al ., ) and MDP (Mitsui et al ., ; Kaboteh et al ., ) and found BSI to contain prognostic and predictive information. Some BSI studies do not describe how the bone scans were performed (Dennis et al ., ; Reza et al ., ; Petersen et al ., ), when images from multiple hospitals were used. Previous studies have compared bone scans performed with different tracers (Bergqvist et al ., ; Fruhling et al ., ).…”

Diagnostic software proposal for bone scan follow-up using false color based on the gammagrams analysis from gray tone histograms

“…In the majority of oncology centers, the gammagram (or bone scan) interpretation is performed visually. However, a vast experience is required to provide an efficient diagnosis, given the difficulties associated with the recognition of hot points (Petersen et al 2017b), like zones with high radiotracer content (Tsui et al 1981 andNordstrom et al 2013a), which is interpreted in the context of gray tones as a concentration of high values or image intensity (Bishnu et al 2005a). From the previous arguments we state that it depends on the scale, reaching values among a maximum of 255 gray ones or even more (Barden et al 1999).…”