Sources of variation in multicenter rectal MRI data and their effect on radiomics feature reproducibility

Objectives To investigate sources of variation in a multicenter rectal cancer MRI dataset focusing on hardware and image acquisition, segmentation methodology, and radiomics feature extraction software. Methods T2W and DWI/ADC MRIs from 649 rectal cancer patients were retrospectively acquired in 9 centers. Fifty-two imaging features (14 first-order/6 shape/32 higher-order) were extracted from each scan using whole-volume (expert/non-expert) and single-slice segmentations using two different software packages (PyRadiomics/CapTk). Influence of hardware, acquisition, and patient-intrinsic factors (age/gender/cTN-stage) on ADC was assessed using linear regression. Feature reproducibility was assessed between segmentation methods and software packages using the intraclass correlation coefficient. Results Image features differed significantly (p < 0.001) between centers with more substantial variations in ADC compared to T2W-MRI. In total, 64.3% of the variation in mean ADC was explained by differences in hardware and acquisition, compared to 0.4% by patient-intrinsic factors. Feature reproducibility between expert and non-expert segmentations was good to excellent (median ICC 0.89–0.90). Reproducibility for single-slice versus whole-volume segmentations was substantially poorer (median ICC 0.40–0.58). Between software packages, reproducibility was good to excellent (median ICC 0.99) for most features (first-order/shape/GLCM/GLRLM) but poor for higher-order (GLSZM/NGTDM) features (median ICC 0.00–0.41). Conclusions Significant variations are present in multicenter MRI data, particularly related to differences in hardware and acquisition, which will likely negatively influence subsequent analysis if not corrected for. Segmentation variations had a minor impact when using whole volume segmentations. Between software packages, higher-order features were less reproducible and caution is warranted when implementing these in prediction models. Key Points • Features derived from T2W-MRI and in particular ADC differ significantly between centers when performing multicenter data analysis. • Variations in ADC are mainly (> 60%) caused by hardware and image acquisition differences and less so (< 1%) by patient- or tumor-intrinsic variations. • Features derived using different image segmentations (expert/non-expert) were reproducible, provided that whole-volume segmentations were used. When using different feature extraction software packages with similar settings, higher-order features were less reproducible.

Objective Sexiness? Using the Statistics of Feature Extraction Algorithm Outcomes to Determine the Sexiness of Images of Nude Women during Pornography Casting

In this paper, we pursue the goal of finding an objective measure of sexiness of the naked female in an (adequately standardized) image of her while she is intent on eliciting sexual desirability (arousal) in the presumed viewer; wecompare it with the objective measure of sexual desirability of her face when posing clothed. The data set consists of images of 15 women posing for selection in a casting (selection) session for participation in pornographic films; we can thus confidently assume that each female wants to appear as sexy as possible. We use AI-based feature extraction algorithms, implemented in software, to analyzefeature vectors of the images. The women’s apparent sexiness can be determined via statistical analysis. We argue that the feature extraction software we use supplies a clustering of these image feature vectors, implying an objective measure of the women’s sexiness. We use such clustering, which implies a software-derived rating, but not ranking, when contrasting the ratings of these images by 50men and 50 women. We argue that, as with the case of a protractor used to measure angles, objectivity can be derived by feature extraction, but not ranking by various human raters. Rather, the statistics of ratings by the men and women inform us of their personal evaluation of sexiness. These statistics of objective sexiness can then be used to infer (statistical) properties of the raters’ personal sexiness perceptions and arousal response against our proposed constructed reference standard.

Longitudinal ellipsoid zone and subretinal fluid mapping following ocriplasmin injection in the prospective observational ORBIT trial

BackgroundOcriplasmin is approved for the treatment of symptomatic vitreomacular traction (VMT). However, several retrospective reports have identified ellipsoid zone (EZ) alterations on spectral domain optical coherence tomography (SDOCT) after ocriplasmin injection. This report quantitatively analysed outer retinal changes after intravitreal ocriplasmin.MethodsOcriplasmin Research to Better Inform Treatment is a prospective, observational phase IV clinical study where subjects received a single intravitreal injection of ocriplasmin for symptomatic VMT. Macular cube scans were imported into a semiautomated EZ mapping and fluid feature extraction software for SDOCT analysis. Change in visual acuity, VMT release, macular hole (MH) closure, EZ integrity/volume and subretinal fluid (SRF) volume on SDOCT macular cube scans were recorded and analysed.ResultsThis analysis included 55 participants with 6 months of follow-up. Intravitreal ocriplasmin injection caused VMT release in 67% and MH closure in 82% of participants. Visual acuity improved by 4.5 letters (p<0.05) in the whole cohort and by 6.0 letters (p<0.05) in participants with VMT release. EZ volume was reduced by 23.4% at week 1 (p<0.001) and recovered to baseline by between months 3 and 6. EZ volume loss at week 1 did not correlate with ETDRS acuity at final visit.ConclusionOcriplasmin treatment resulted in VMT release, MH closure and visual acuity gains in a significant portion of eyes. EZ volume was significantly reduced at week 1, but recovered to baseline levels by final follow-up and was not associated with final visual acuity.