A B S T R A C T Precisely identifying sulcal features in brain MRI is made challenging by the variability of brain folding. This research introduces an innovative 3D explainability frame-work that validates outputs from deep learning networks in their ability to detect the paracin-gulate sulcus, an anatomical feature that may or may not be present on the frontal medial surface of the human brain. This study trained and tested two networks, amalgamating local explainability techniques GradCam and SHAP with a dimensionality reduction method. The explainability framework provided both localized and global explanations, along with accuracy of classification results, revealing pertinent sub-regions contributing to the decision process through a post-fusion transformation of explanatory and statistical features. Leveraging the TOP-OSLO dataset of MRI acquired from patients with schizophrenia, greater accuracies of paracingulate sulcus detection (presence or absence) were found in the left compared to right hemispheres with distinct, but extensive sub-regions contributing to each classification outcome. The study also inadvertently highlighted the critical role of an unbiased annotation protocol in maintaining network performance fairness. Our proposed method not only o ff ers automated, impartial annotations of a variable sulcus but also provides insights into the broader anatomical variations associated with its presence throughout the brain. The adoption of this methodology holds promise for instigating further explorations and inquiries in the field of neuroscience.1. Introduction While the folding of the primary sulci of the human brain, formed during gestation, is broadly stable across individuals, the secondary sulci which continue to develop post-natally are unique to each individual. Inter-individual variability poses a significant challenge for the detection and accurately annotation of sulcal features from MRI of the brain. Undertaking this task manually is time-consuming with outcomes that depend on the rater. This prevents the e fficient leveraging of the large, open-access MRI databases that are available. While primary sulci can be very accurately detected with automated methods, secondary sulci pose a more di fficult computational problem due to their higher variability in shape and indeed presence or absense [3]. A successful automated method would facilitate investigations of brain folding variation, representative of events occurring during a critical developmental period. Furthermore, generalized and unbiased annotations would make tractable large-scale studies of cognitive and behavioral development, and the emergence of mental and neurological disorders with high levels of statistical power. The folding of the brain has been linked to brain function, and some specific folding patterns have been related to susceptibility to neurological adversities [20].

In 1979, when Steven Spielberg and George Lucas signed with Paramount Pictures to develop a film series based on classic Hollywood adventure serials, the deal they struck was to make five separate movies. The first, Raiders of the Lost Ark, was released in the summer of 1981 and became that year's top-grossing movie, beating even the long-anticipated Superman II and remaining on screens in some cities for more than a year. By 1984, it was Raiders' sequel, Indiana Jones and the Temple of Doom, that had become the year's most anticipated movie, banking that year's biggest opening weekend, and the third entry in the franchise, Indiana Jones and the Last Crusade, became not only the highest-grossing movie of 1989 but the top-earning Indiana Jones movie yet. Given that track record, and viewed from the perspective of our own IP-crazed times, it seems inconceivable that Spielberg and Lucas decided not to move forward immediately with a fourth Indiana Jones picture (though Lucas did go on to produce a spinoff TV series, The Young Indiana Jones Chronicles). Lucas' idea, a riff on the 1950s sci-fi films that would have been contemporaneous with a middle-aged Indy, was to introduce extraterrestrial beings into the previously earthbound Raiders universe.

The cerebellum has long been recognized as a partner of the cerebral cortex, and both have expanded greatly in human evolution. The thin cerebellar cortex is even more tightly folded than the cerebral cortex. By scanning a human cerebellum specimen at ultra-high magnetic fields, we were able to computationally reconstruct its surface down to the level of the smallest folds, revealing that the cerebellar cortex has almost 80% of the surface area of the cerebral cortex. By performing the same procedure on a monkey brain, we found that the surface area of the human cerebellum has expanded even more than that of the human cerebral cortex, suggesting a role in characteristically human behaviors, such as toolmaking and language. The surface of the human cerebellar cortex is much more tightly folded than the cerebral cortex. It was computationally reconstructed for the first time to the level of all individual folia from multicontrast high-resolution postmortem MRI scans. Its total shrinkage-corrected surface area (1,590 cm2) was larger than expected or previously reported, equal to 78% of the total surface area of the human neocortex. The unfolded and flattened surface comprised a narrow strip 10 cm wide but almost 1 m long. By applying the same methods to the neocortex and cerebellum of the macaque monkey, we found that its cerebellum was relatively much smaller, approximately 33% of the total surface area of its neocortex. This suggests a prominent role for the cerebellum in the evolution of distinctively human behaviors and cognition. Datasets including (i) original high-resolution isotropic 3D MRI data of the human cerebellum with two different contrasts; (ii) computationally combined, normalized, filtered, and edited versions of that 3D data; and (iii) FreeSurfer-compatible subject surfaces and vertexwise measurements reconstructed from the data can be downloaded from . Software for performing the analyses presented in this paper is available at or [http://www.cogsci.ucsd.edu/∼sereno/.tmp/dist/csurf][1]. We also used some of the utilities from the standard FreeSurfer 5.3 distribution (available at [https://surfer.nmr.mgh.harvard.edu][2]/) and from the AFNI distribution (available at [https://afni.nimh.nih.gov][3]/). [1]: http://www.cogsci.ucsd.edu/%7Esereno/.tmp/dist/csurf [2]: https://surfer.nmr.mgh.harvard.edu/ [3]: https://afni.nimh.nih.gov/

Pulmonary lobe segmentation is an important task for pulmonary disease related Computer Aided Diagnosis systems (CADs). Classical methods for lobe segmentation rely on successful detection of fissures and other anatomical information such as the location of blood vessels and airways. With the success of deep learning in recent years, Deep Convolutional Neural Network (DCNN) has been widely applied to analyze medical images like Computed Tomography (CT) and Magnetic Resonance Imaging (MRI), which, however, requires a large number of ground truth annotations. In this work, we release our manually labeled 50 CT scans which are randomly chosen from the LUNA16 dataset and explore the use of deep learning on this task. We propose pre-processing CT image by cropping region that is covered by the convex hull of the lungs in order to mitigate the influence of noise from outside the lungs. Moreover, we design a hybrid loss function with dice loss to tackle extreme class imbalance issue and focal loss to force model to focus on voxels that are hard to be discriminated. To validate the robustness and performance of our proposed framework trained with a small number of training examples, we further tested our model on CT scans from an independent dataset. Experimental results show the robustness of the proposed approach, which consistently improves performance across different datasets by a maximum of $5.87\%$ as compared to a baseline model.

The Arizona Geological Survey is monitoring a 2-mile long crack that has opened up in the Arizona desert. Recent drone flights over the crack reveal that it has continued to grow both in length and width in Pinal County, to the southeast of Phoenix. Scientists are actively monitoring the crack and took drone video of the extent of the fissure as normal documentation of an area prone to large cracks in the Earth. The northern portion of the crack is older and partially filled in by eroding sediment and from collapse of the crack's edges. Meanwhile, the southern portion remains 25 to 30 feet deep and 10 feet across.

A huge two mile-long crack has been discovered in the desert in Arizona. Drone footage uploaded to YouTube by the Arizona Geological Survey shows the massive fissure splitting the desert's surface in the Tator Hills area of southern Pinal County. The film shows people dwarfed by the crack as they stand next to the edge, while the drone flies over the wide-open fissure which extends farther into the earth then the eye can see. It is the first time that the AZGS has used drone footage to examine fissures in this way. 'AZGS is experimenting with drone technology as a tool for mapping fissures and other surface features, e.g.

A fissure almost two miles long has been discovered in the Arizona desert. Drone footage uploaded to YouTube by the Arizona Geological Survey shows the massive fissure in the desert surface. In a post accompanying the video the Arizona Geological Survey explains that the fissure is in the Tator Hills area of southern Pinal County. "AZGS is experimenting with drone technology as a tool for mapping fissures and other surface features, e.g. Joe Cook of the Arizona Geological Survey told 12News that the fissure's southern mile is fresher, and may have reached the surface in a 2016 monsoon.