Do we need the third dimension? Quantifying the effect of the z-axis in 3D geometric morphometrics based on sailfin silversides (Telmatherinidae).

This study investigated the impact of the third dimension in geometric morphometrics (GM) using sailfin silversides (Telmatherinidae) from the Malili Lakes of Sulawesi (Indonesia). The three morphospecies of the monophyletic "roundfin" radiation are laterally compressed and vary in shape traits. The results of 2D and 3D GM were compared and quantified to discuss the advantages and disadvantages of both methods for closely related species and their sexes. This approach focused on the head because it is far more complex and three-dimensionally structured than the trunk or the caudal region. The results revealed no significant benefit concerning repeatability and measurement error in 3D GM compared to 2D GM. The z-axis contributed substantially to the variance of the 3D data set but was irrelevant for discrimination of species and sexes in the approach. Limited gain in information was contrasted by substantially higher effort for 3D compared to the 2D analyses. The study concluded that 2D GM is the more efficient shape analysis approach for discriminating roundfins. Broader studies are needed to test which of the two methods is more efficient in distinguishing laterally compressed fishes in general. For future studies, due to the high investment required, this study recommends carefully evaluating the necessity of 3D GM. If in doubt, this study suggests testing for congruence between 2D and 3D GM with a subsample and consequently applying 2D GM in the case of high congruence.

Three-Dimensional Analysis of Component Stability of the Nellix Endovascular Aneurysm Sealing System After Treatment of Infrarenal Abdominal Aortic Aneurysms.

PURPOSE: To assess short-term stability and conformational changes of the Nellix EndoVascular Aneurysm Sealing (EVAS) System using 3-dimensional (3D) analysis. METHODS: Postoperative computed tomography (CT) scans obtained at 0, 3, and 12 months in 24 patients (mean age 75±7 years; 22 men) who underwent EVAS between December 2013 and December 2014 for intact abdominal aortic aneurysm (within the instructions for use) were evaluated for stent-graft deviation in multiple planes using dedicated 3D analysis software. In addition, 2D analysis using an anatomically fixed reference landmark was performed to assess craniocaudal migration. Clinical and follow-up data of the patients were recorded and matched with results of the imaging analysis. RESULTS: Overall stability of the Nellix endografts was promising. Relevant conformational changes in the majority of cases were limited to the iliac graft segment and were clinically benign in all cases. Conversely, the only deviation of the proximal stent-graft segment was found in a patient with type Ia endoleak. Additional 2D analysis found relevant (≥5 mm) caudal migration of the Nellix stent-graft in 6 patients, including the one with the type Ia endoleak. In 3 patients, 3D analysis demonstrated the absence of relevant conformational changes of the endografts despite caudal migration. CONCLUSION: Overall stability of the separate EVAS stent-grafts is promising in the short term. Relevant conformational changes (stent-graft deviation) in the majority of cases were benign and confined to the iliac segment.

Machine learning prediction of axillary lymph node metastasis in breast cancer: 2D versus 3D radiomic features.

PURPOSE: The purpose of this study was to distinguish axillary lymph node (ALN) status using preoperative breast DCE-MRI radiomics and compare the effects of two-dimensional (2D) and three-dimensional (3D) analysis. METHODS: A retrospective study including 154 breast cancer patients all confirmed by pathology; 80 with ALN metastasis and 74 without. All MRI scans were achieved at a 3.0 Tesla scanner with 7 post-contrast MR phases sequentially acquired with a temporal resolution of 60 s. MRI radiomic features were extracted separately from a 2D single slice (i.e., the representative slice) and the 3D tumor volume. Several machine learning classifiers were built and compared using 2D or 3D analysis to distinguish positive vs negative ALN status. We performed independent test and 10-fold cross validation with multiple repetitions, and used bootstrap test, least absolute shrinkage selection operator, and receiver operating characteristic (ROC) curve analysis as statistical tests. RESULTS: The highest area under the ROC curve (AUC) was 0.81 (95% confidence intervals [CI]: 0.80-0.83) and 0.82 (95% CI: 0.81-0.82) for 2D and 3D analysis, respectively; the corresponding accuracy was 79% and 80%. The linear discriminant analysis (LDA) classifier achieved the highest classification performance. None of the AUC differences between 2D and 3D analysis was statistically significant for the several tested machine learning classifiers (all P> 0.05). CONCLUSIONS: Radiomic features from segmented tumor region in breast MRI were associated with ALN status. The separate radiomic analysis on 3D tumor volume showed a similar effect to the 2D analysis on the single representative slice in the tested machine learning classifiers.