Radiomics based automated quality assessment for T2W prostate MR images.

PURPOSE: The guidelines for prostate cancer recommend the use of MRI in the prostate cancer pathway. Due to the variability in prostate MR image quality, the reliability of this technique in the detection of prostate cancer is highly variable in clinical practice. This leads to the need for an objective and automated assessment of image quality to ensure an adequate acquisition and hereby to improve the reliability of MRI. The aim of this study is to investigate the feasibility of Blind/referenceless image spatial quality evaluator (Brisque) and radiomics in automated image quality assessment of T2-weighted (T2W) images. METHOD: Anonymized axial T2W images from 140 patients were scored for quality using a five-point Likert scale (low, suboptimal, acceptable, good, very good quality) in consensus by two readers. Images were dichotomized into clinically acceptable (very good, good and acceptable quality images) and clinically unacceptable (low and suboptimal quality images) in order to train and verify the model. Radiomics and Brisque features were extracted from a central cuboid volume including the prostate. A reduced feature set was used to fit a Linear Discriminant Analysis (LDA) model to predict image quality. Two hundred times repeated 5-fold cross-validation was used to train the model and test performance by assessing the classification accuracy, the discrimination accuracy as receiver operating curve - area under curve (ROC-AUC), and by generating confusion matrices. RESULTS: Thirty-four images were classified as clinically unacceptable and 106 were classified as clinically acceptable. The accuracy of the independent test set (mean ± standard deviation) was 85.4 ± 5.5%. The ROC-AUC was 0.856 (0.851 - 0.861) (mean; 95% confidence interval). CONCLUSIONS: Radiomics AI can automatically detect a significant portion of T2W images of suboptimal image quality. This can help improve image quality at the time of acquisition, thus reducing repeat scans and improving diagnostic accuracy.

Cost-effectiveness of artificial intelligence aided vessel occlusion detection in acute stroke: an early health technology assessment.

BACKGROUND: Limited evidence is available on the clinical impact of artificial intelligence (AI) in radiology. Early health technology assessment (HTA) is a methodology to assess the potential value of an innovation at an early stage. We use early HTA to evaluate the potential value of AI software in radiology. As a use-case, we evaluate the cost-effectiveness of AI software aiding the detection of intracranial large vessel occlusions (LVO) in stroke in comparison to standard care. We used a Markov based model from a societal perspective of the United Kingdom predominantly using stroke registry data complemented with pooled outcome data from large, randomized trials. Different scenarios were explored by varying missed diagnoses of LVOs, AI costs and AI performance. Other input parameters were varied to demonstrate model robustness. Results were reported in expected incremental costs (IC) and effects (IE) expressed in quality adjusted life years (QALYs). RESULTS: Applying the base case assumptions (6% missed diagnoses of LVOs by clinicians, $40 per AI analysis, 50% reduction of missed LVOs by AI), resulted in cost-savings and incremental QALYs over the projected lifetime (IC: - $156, - 0.23%; IE: + 0.01 QALYs, + 0.07%) per suspected ischemic stroke patient. For each yearly cohort of patients in the UK this translates to a total cost saving of $11 million. CONCLUSIONS: AI tools for LVO detection in emergency care have the potential to improve healthcare outcomes and save costs. We demonstrate how early HTA may be applied for the evaluation of clinically applied AI software for radiology.

Cost-Effectiveness Comparison of Imaging-Guided Prostate Biopsy Techniques: Systematic Transrectal Ultrasound, Direct In-Bore MRI, and Image Fusion.

OBJECTIVE: Three commonly used prostate biopsy approaches are systematic transrectal ultrasound guided, direct in-bore MRI guided, and image fusion guided. The aim of this study was to calculate which strategy is most cost-effective. MATERIALS AND METHODS: A decision tree and Markov model were developed to compare cost-effectiveness. Literature review and expert opinion were used as input. A strategy was deemed cost-effective if the costs of gaining one quality-adjusted life year (incremental cost-effectiveness ratio) did not exceed the willingness-to-pay threshold of €80,000 (≈$85,000 in January 2017). A base case analysis was performed to compare systematic transrectal ultrasound- and image fusion-guided biopsies. Because of a lack of appropriate literature regarding the accuracy of direct in-bore MRI-guided biopsy, a threshold analysis was performed. RESULTS: The incremental cost-effectiveness ratio for fusion-guided biopsy compared with systematic transrectal ultrasound-guided biopsy was €1386 ($1470) per quality-adjusted life year gained, which was below the willingness-to-pay threshold and thus assumed cost-effective. If MRI findings are normal in a patient with clinically significant prostate cancer, the sensitivity of direct in-bore MRI-guided biopsy has to be at least 88.8%. If that is the case, the incremental cost-effectiveness ratio is €80,000 per quality-adjusted life year gained and thus cost-effective. CONCLUSION: Fusion-guided biopsy seems to be cost-effective compared with systematic transrectal ultrasound-guided biopsy. Future research is needed to determine whether direct in-bore MRI-guided biopsy is the best pathway; in this study a threshold was calculated at which it would be cost-effective.

Cost-effectiveness of magnetic resonance (MR) imaging and MR-guided targeted biopsy versus systematic transrectal ultrasound-guided biopsy in diagnosing prostate cancer: a modelling study from a health care perspective.

BACKGROUND: The current diagnostic strategy using transrectal ultrasound-guided biopsy (TRUSGB) raises concerns regarding overdiagnosis and overtreatment of prostate cancer (PCa). Interest in integrating multiparametric magnetic resonance imaging (MRI) and magnetic resonance-guided biopsy (MRGB) into the diagnostic pathway to reduce overdiagnosis and improve grading is gaining ground, but it remains uncertain whether this image-based strategy is cost-effective. OBJECTIVE: To determine the cost-effectiveness of multiparametric MRI and MRGB compared with TRUSGB. DESIGN, SETTING, AND PARTICIPANTS: A combined decision tree and Markov model for men with elevated prostate-specific antigen (>4 ng/ml) was developed. Input data were derived from systematic literature searches, meta-analyses, and expert opinion. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: Quality-adjusted life years (QALYs) and health care costs of both strategies were modelled over 10 yr after initial suspicion of PCa. Probabilistic and threshold analyses were performed to assess uncertainty. RESULTS AND LIMITATIONS: Despite uncertainty around the presented cost-effectiveness estimates, our results suggest that the MRI strategy is cost-effective compared with the standard of care. Expected costs per patient were € 2423 for the MRI strategy and € 2392 for the TRUSGB strategy. Corresponding QALYs were higher for the MRI strategy (7.00 versus 6.90), resulting in an incremental cost-effectiveness ratio of € 323 per QALY. Threshold analysis revealed that MRI is cost-effective when sensitivity of MRGB is ≥ 20%. The probability that the MRI strategy is cost-effective is around 80% at willingness to pay thresholds higher than € 2000 per QALY. CONCLUSIONS: Total costs of the MRI strategy are almost equal with the standard of care, while reduction of overdiagnosis and overtreatment with the MRI strategy leads to an improvement in quality of life. PATIENT SUMMARY: We compared costs and quality of life (QoL) of the standard "blind" diagnostic technique with an image-based technique for men with suspicion of prostate cancer. Our results suggest that costs were comparable, with higher QoL for the image-based technique.

Prostate cancer aggressiveness: in vivo assessment of MR spectroscopy and diffusion-weighted imaging at 3 T.

PURPOSE: To determine the individual and combined performance of magnetic resonance (MR) spectroscopic imaging and diffusion-weighted (DW) imaging at 3 T in the in vivo assessment of prostate cancer aggressiveness by using histopathologically defined regions of interest on radical prostatectomy specimens to define the prostate cancer regions to be investigated. MATERIALS AND METHODS: The local institutional ethics review board approved this retrospective study and waived the informed consent requirement. Fifty-four patients with biopsy-proved prostate cancer underwent clinical MR spectroscopic imaging followed by prostatectomy. Guided by the histopathologic map, all spectroscopy voxels that contained tumor tissue were selected, and metabolite ratios (choline [Cho] plus creatine [Cr]-to-citrate [Cit] and Cho/Cr ratios) were derived. For each spectroscopic voxel, 25th percentile apparent diffusion coefficient (ADC) of the region corresponding to that voxel was determined, representing the most aberrant tumor part on the ADC map, which was often smaller than spectroscopic imaging voxels. Maximum metabolic ratios and minimum 25th percentile ADC of each tumor were related to tumor aggressiveness and were used to differentiate aggressiveness classes. A logistic regression model (LRM) was used to combine data from both modalities. RESULTS: Significant correlation was found between aggressiveness classes and maximum Cho+Cr/Cit ratio (ρ=0.36), maximum Cho/Cr ratio (ρ=0.35), and minimum 25th percentile ADC (ρ=-0.63) in the peripheral zone (PZ). In the transition zone (TZ), the correlation was significant for only Cho+Cr/Cit and Cho/Cr ratios (ρ=0.58 and ρ=0.60, respectively). For differentiation between aggressiveness classes, LRM use did not result in significantly improved differentiation over any individual variables. CONCLUSION: These findings enabled confirmation that MR spectroscopic imaging and DW imaging offer potential for in vivo noninvasive assessment of prostate cancer aggressiveness, and both modalities have comparable performance. The combination did not result in better performance. Nonetheless, the better performances of metabolite ratios in the TZ and of ADCs in the PZ suggest that they have complementary value.