Expert System for Bone Scan Interpretation Improves Progression Assessment in Bone Metastatic Prostate Cancer.

Haupt, Fabian; Berding, Georg; Namazian, Ali; Wilke, Florian; Böker, Alena; Merseburger, Axel; Geworski, Lilli; Kuczyk, Markus Antonius; Bengel, Frank Michael; Peters, Inga (2017). Expert System for Bone Scan Interpretation Improves Progression Assessment in Bone Metastatic Prostate Cancer. Advances in therapy, 34(4), pp. 986-994. Springer 10.1007/s12325-017-0505-z

[img] Text
Expert System for Bone Scan Interpretation Improves Progression Assessment in Bone Metastatic Prostate Cancer..pdf - Published Version
Restricted to registered users only
Available under License Publisher holds Copyright.

Download (895kB) | Request a copy

INTRODUCTION The bone scan index (BSI) was introduced as a quantitative tool for tumor involvement in bone of patients with metastatic prostate cancer (mPCa). The computer-aided diagnosis device for BSI analysis EXINIbone(BSI) seems to represent technical progress for the quantitative assessment of bone involvement. But it is not yet clear if the automated BSI (aBSI) could contribute to improved evaluation of progression in patients under antiandrogens or chemotherapy in contrast to the visual interpretation and/or conventional biomarkers such as the prostate-specific antigen (PSA). METHODS In 49 mPCa patients, bone scans were performed initially and during different therapy courses. Scans were evaluated visually and by the artificial-neural-network-based expert system EXINIbone(BSI). Progression of metastatic bone involvement was defined according to the Prostate Cancer Clinical Trials Working Group 2 (PCWG2) criteria in the visual interpretation. The computer-assisted interpretation was based on different cutoff values in relative changes of the aBSI. Additionally, assessments according to bone scanning were compared to changes in the PSA value as a potential surrogate for treatment response. RESULTS Using a sensitive cutoff value (5% or 10%) for the relative aBSI increase led to significantly increased progression determination compared to the visual interpretation of bone scans (49% and 43% vs. 27%, p < 0.001). In 63% of the cases PSA and BSI changes matched, whereas in 18% progression was only indicated by the aBSI. A relative cutoff of 5% for the aBSI decrease could reclassify 47 serial scan pairs which were visually interpreted as stable into 22 progressive and 25 remissive scans. CONCLUSION Distinct thresholds of the relative aBSI could help to better assess disease progression in mPCa patients. Manual corrections of the BSI values are not required in most cases. The aBSI could serve as a useful additional parameter for therapy monitoring in mPCa patients in the future.

Item Type:

Journal Article (Original Article)


04 Faculty of Medicine > Department of Radiology, Neuroradiology and Nuclear Medicine (DRNN) > Clinic of Nuclear Medicine

UniBE Contributor:

Haupt, Fabian


600 Technology > 610 Medicine & health








Franziska Nicoletti

Date Deposited:

08 Aug 2017 17:29

Last Modified:

20 Mar 2018 13:23

Publisher DOI:


PubMed ID:


Uncontrolled Keywords:

Bone scan index (BSI); Bone scintigraphy; CRPC; Computer-aided diagnosis device; Metastatic prostate cancer; Therapy response




Actions (login required)

Edit item Edit item
Provide Feedback