Radiology
-
The use of artificial intelligence in medicine is currently an issue of great interest, especially with regard to the diagnostic or predictive analysis of medical images. Adoption of an artificial intelligence tool in clinical practice requires careful confirmation of its clinical utility. ⋯ Next, the effects of disease manifestation spectrum and disease prevalence on the performance results are explained, followed by a discussion of the difference between evaluating the performance with use of internal and external datasets, the importance of using an adequate external dataset obtained from a well-defined clinical cohort to avoid overestimating the clinical performance as a result of overfitting in high-dimensional or overparameterized classification model and spectrum bias, and the essentials for achieving a more robust clinical evaluation. Finally, the authors review the role of clinical trials and observational outcome studies for ultimate clinical verification of diagnostic or predictive artificial intelligence tools through patient outcomes, beyond performance metrics, and how to design such studies. © RSNA, 2018.
-
Controlled Clinical Trial
Middle Meningeal Artery Embolization for Chronic Subdural Hematoma.
Purpose To evaluate the effect of middle meningeal artery (MMA) embolization on chronic subdural hematoma (CSDH) and compare the treatment outcomes of MMA embolization and conventional treatment. Materials and Methods All consecutive patients 20 years or older with CSDH were assessed for eligibility. CSDHs with a focal location, a thickness of 10 mm or less, no mass effect, or underlying conditions were excluded. ⋯ Surgical rescue was less frequent in the embolization group (one of 72 patients [1.4%] vs 88 of 469 patients [18.8%]; adjusted OR, 0.094; 95% CI: 0.018, 0.488; P = .005). Treatment-related complication rate was not different between the two groups (0 of 72 patients vs 20 of 469 patients [4.3%]; adjusted OR, 0.145; 95% CI: 0.009, 2.469; P = .182). Conclusion MMA embolization has a positive therapeutic effect on CSDH and is more effective than conventional treatment. © RSNA, 2017.
-
Purpose To evaluate whether arterial input functions (AIFs) derived from dynamic susceptibility-contrast (DSC) magnetic resonance (MR) imaging, or AIFDSC values, improve diagnostic accuracy and reliability of the pharmacokinetic (PK) parameters of dynamic contrast material-enhanced (DCE) MR imaging for differentiating high-grade from low-grade astrocytomas, compared with AIFs obtained from DCE MR imaging (AIFDCE). Materials and Methods This retrospective study included 226 patients (138 men, 88 women; mean age, 52.27 years ± 15.17; range, 24-84 years) with pathologically confirmed astrocytomas (World Health Organization grade II = 21, III = 53, IV = 152; isocitrate dehydrogenase mutant, 11.95% [27 of 226]; 1p19q codeletion 0% [0 of 226]). All patients underwent both DSC and DCE MR imaging before surgery, and AIFDSC and AIFDCE were obtained from each image. ⋯ All three parameters had better ICCs with AIFDSC than with AIFDCE (Ktrans, 0.737 vs 0.095; vp, 0.848 vs 0.728; ve, 0.875 vs 0.581, respectively). In AIF analysis, maximal signal intensity (0.837 vs 0.524) and wash-in slope (0.800 vs 0.432) demonstrated better ICCs with AIFDSC than AIFDCE. Conclusion AIFDSC-driven DCE MR imaging PK parameters showed better diagnostic accuracy and reliability for differentiating high-grade from low-grade astrocytoma than those derived from AIFDCE. © RSNA, 2017 Online supplemental material is available for this article.
-
Purpose To correlate quantitative diffusion-weighted imaging (DWI) parameters derived from conventional monoexponential DWI, stretched exponential DWI, diffusion kurtosis imaging (DKI), and diffusion-tensor imaging (DTI) with quantitative histopathologic tumor tissue composition in prostate cancer in a preliminary hypothesis-generating study. Materials and Methods This retrospective institutional review board-approved study included 24 patients with prostate cancer (mean age, 63 years) who underwent magnetic resonance (MR) imaging, including high-b-value DWI and DTI at 3.0 T, before prostatectomy. The following parameters were calculated in index tumors and nontumoral peripheral zone (PZ): apparent diffusion coefficient (ADC) obtained with monoexponential fit (ADCME), ADC obtained with stretched exponential modeling (ADCSE), anomalous exponent (α) obtained at stretched exponential DWI, ADC obtained with DKI modeling (ADCDKI), kurtosis with DKI, ADC obtained with DTI (ADCDTI), and fractional anisotropy (FA) at DTI. ⋯ These findings should be validated in a larger study. © RSNA, 2017 Online supplemental material is available for this article. An earlier incorrect version of this article appeared online. This article was corrected on November 10, 2017.
-
Purpose To compare the diagnostic yield and complication rates of electromagnetic navigational bronchoscopic (ENB)-guided and computed tomography (CT)-guided percutaneous tissue sampling of lung nodules. Materials and Methods Retrospectively identified were 149 patients sampled percutaneously with CT guidance and 146 patients who underwent ENB with transbronchial biopsy of a lung lesion between 2013 and 2015. Clinical data, incidence of complications, and nodule pathologic analyses were assessed through electronic medical record review. ⋯ Similar yield for molecular analyses was noted with the two approaches (ENB-guided sampling, 88.9% [32 of 36]; CT-guided sampling, 82.0% [41 of 50]). The two groups had similar rates of major complications (symptomatic hemorrhage, P > .999; pneumothorax requiring chest tube and/or admission, P = .417). Conclusion CT-guided transthoracic biopsy provided higher diagnostic yield in the assessment of peripheral pulmonary nodules than navigational bronchoscopy with a similar rate of clinically relevant complications. © RSNA, 2017 Online supplemental material is available for this article.