Algorithms for computer-aided diagnosis of dementia based on structural MRI have demonstrated high performance in the literature, but are difficult to compare as different data sets and methodology were used for evaluation. In addition, it is unclear how the algorithms would perform on previously unseen data, and thus, how they would perform in clinical practice when there is no real opportunity to adapt the algorithm to the data at hand. To address these comparability, generalizability and clinical applicability issues, we organized a grand challenge that aimed to objectively compare algorithms based on a clinically representative multi-center data set. Using clinical practice as the starting point, the goal was to reproduce the clinical diagnosis. Therefore, we evaluated algorithms for multi-class classification of three diagnostic groups: patients with probable Alzheimer's disease, patients with mild cognitive impairment and healthy controls. The diagnosis based on clinical criteria was used as reference standard, as it was the best available reference despite its known limitations. For evaluation, a previously unseen test set was used consisting of 354 T1-weighted MRI scans with the diagnoses blinded. Fifteen research teams participated with a total of 29 algorithms. The algorithms were trained on a small training set (n=30) and optionally on data from other sources (e.g., the Alzheimer's Disease Neuroimaging Initiative, the Australian Imaging Biomarkers and Lifestyle flagship study of aging). The best performing algorithm yielded an accuracy of 63.0% and an area under the receiver-operating-characteristic curve (AUC) of 78.8%. In general, the best performances were achieved using feature extraction based on voxel-based morphometry or a combination of features that included volume, cortical thickness, shape and intensity. The challenge is open for new submissions via the web-based framework: http://caddementia.grand-challenge.org.

Imaging using 3-D DESI mass spectral data combined with deep learning reveals the topology and heterogeneity of colorectal cancer.

Identifying accurate biomarkers of cognitive decline is essential for advancing early diagnosis and prevention therapies in Alzheimer's disease. The Alzheimer's disease DREAM Challenge was designed as a computational crowdsourced project to benchmark the current state‐of‐the‐art in predicting cognitive outcomes in Alzheimer's disease based on high dimensional, publicly available genetic and structural imaging data. This meta‐analysis failed to identify a meaningful predictor developed from either data modality, suggesting that alternate approaches should be considered for prediction of cognitive performance.

The pregnancy vaginal microbiome contributes to risk of preterm birth, the primary cause of death in children under 5 years of age. Here we describe direct on-swab metabolic profiling by Desorption Electrospray Ionization Mass Spectrometry (DESI-MS) for sample preparation-free characterisation of the cervicovaginal metabolome in two independent pregnancy cohorts (VMET, n = 160; 455 swabs; VMET II, n = 205; 573 swabs). By integrating metataxonomics and immune profiling data from matched samples, we show that specific metabolome signatures can be used to robustly predict simultaneously both the composition of the vaginal microbiome and host inflammatory status. In these patients, vaginal microbiota instability and innate immune activation, as predicted using DESI-MS, associated with preterm birth, including in women receiving cervical cerclage for preterm birth prevention. These findings highlight direct on-swab metabolic profiling by DESI-MS as an innovative approach for preterm birth risk stratification through rapid assessment of vaginal microbiota-host dynamics.

Tumors are intrinsically heterogeneous and it is well established that this directs their evolution, hinders their classification and frustrates therapy^1–3. Consequently, spatially resolved omics-level analyses are gaining traction^4–9. Despite considerable therapeutic interest, tumor metabolism has been lagging behind this development and there is a paucity of data regarding its spatial organization. To address this shortcoming, we set out to study the local metabolic effects of the oncogene c-MYC, a pleiotropic transcription factor that accumulates with tumor progression and influences metabolism^10,11. Through correlative mass spectrometry imaging, we show that pantothenic acid (vitamin B₅) associates with MYC-high areas within both human and murine mammary tumors, where its conversion to coenzyme A fuels Krebs cycle activity. Mechanistically, we show that this is accomplished by MYC-mediated upregulation of its multivitamin transporter SLC5A6. Notably, we show that SLC5A6 over-expression alone can induce increased cell growth and a shift toward biosynthesis, whereas conversely, dietary restriction of pantothenic acid leads to a reversal of many MYC-mediated metabolic changes and results in hampered tumor growth. Our work thus establishes the availability of vitamins and cofactors as a potential bottleneck in tumor progression, which can be exploited therapeutically. Overall, we show that a spatial understanding of local metabolism facilitates the identification of clinically relevant, tractable metabolic targets.