Artificial intelligence for multimodal data integration in oncology.

In oncology, the patient state is characterized by a whole spectrum of modalities, ranging from radiology, histology, and genomics to electronic health records. Current artificial intelligence (AI) models operate mainly in the realm of a single modality, neglecting the broader clinical context, which inevitably diminishes their potential. Integration of different data modalities provides opportunities to increase robustness and accuracy of diagnostic and prognostic models, bringing AI closer to clinical practice. AI models are also capable of discovering novel patterns within and across modalities suitable for explaining differences in patient outcomes or treatment resistance. The insights gleaned from such models can guide exploration studies and contribute to the discovery of novel biomarkers and therapeutic targets. To support these advances, here we present a synopsis of AI methods and strategies for multimodal data fusion and association discovery. We outline approaches for AI interpretability and directions for AI-driven exploration through multimodal data interconnections. We examine challenges in clinical adoption and discuss emerging solutions.

High-resolution structural and functional deep brain imaging using adaptive optics three-photon microscopy.

Multiphoton microscopy has become a powerful tool with which to visualize the morphology and function of neural cells and circuits in the intact mammalian brain. However, tissue scattering, optical aberrations and motion artifacts degrade the imaging performance at depth. Here we describe a minimally invasive intravital imaging methodology based on three-photon excitation, indirect adaptive optics (AO) and active electrocardiogram gating to advance deep-tissue imaging. Our modal-based, sensorless AO approach is robust to low signal-to-noise ratios as commonly encountered in deep scattering tissues such as the mouse brain, and permits AO correction over large axial fields of view. We demonstrate near-diffraction-limited imaging of deep cortical spines and (sub)cortical dendrites up to a depth of 1.4 mm (the edge of the mouse CA1 hippocampus). In addition, we show applications to deep-layer calcium imaging of astrocytes, including fibrous astrocytes that reside in the highly scattering corpus callosum.

Data-efficient and weakly supervised computational pathology on whole-slide images.

Deep-learning methods for computational pathology require either manual annotation of gigapixel whole-slide images (WSIs) or large datasets of WSIs with slide-level labels and typically suffer from poor domain adaptation and interpretability. Here we report an interpretable weakly supervised deep-learning method for data-efficient WSI processing and learning that only requires slide-level labels. The method, which we named clustering-constrained-attention multiple-instance learning (CLAM), uses attention-based learning to identify subregions of high diagnostic value to accurately classify whole slides and instance-level clustering over the identified representative regions to constrain and refine the feature space. By applying CLAM to the subtyping of renal cell carcinoma and non-small-cell lung cancer as well as the detection of lymph node metastasis, we show that it can be used to localize well-known morphological features on WSIs without the need for spatial labels, that it overperforms standard weakly supervised classification algorithms and that it is adaptable to independent test cohorts, smartphone microscopy and varying tissue content.

Enhancing Interpretability of Gene Signatures with Prior Biological Knowledge.

Biological interpretability is a key requirement for the output of microarray data analysis pipelines. The most used pipeline first identifies a gene signature from the acquired measurements and then uses gene enrichment analysis as a tool for functionally characterizing the obtained results. Recently Knowledge Driven Variable Selection (KDVS), an alternative approach which performs both steps at the same time, has been proposed. In this paper, we assess the effectiveness of KDVS against standard approaches on a Parkinson's Disease (PD) dataset. The presented quantitative analysis is made possible by the construction of a reference list of genes and gene groups associated to PD. Our work shows that KDVS is much more effective than the standard approach in enhancing the interpretability of the obtained results.

Land management versus natural factors in land instability: some examples in northern Spain.

The objective of this work is to test a hypothesis formulated on the basis of former results which considers that there might be a ''global geomorphic change,'' due to activities related to land management and not determined by climate change, which could be causing an acceleration of geomorphic processes. Possible relationships between some geomorphic processes related to land instability (landslides or sediment generation) and potential triggering factors are analyzed in study areas in northern Spain. The analysis is based on landslide inventories covering different periods, as well as the determination of sedimentation rates. Temporal landslide and sedimentation rate trends are compared with different indicators of human activities (land-use change, logging, forest fires) and with potential natural triggers (rainfall, seismicity). The possible influence of the road network in the distribution of landslides is also analyzed. Results obtained show that there is a general increase of both landslide and sedimentation rates with time that cannot be explained satisfactorily by observed rainfall trends and even less by seismicity. Land use change appears to be by far the main factor leading to land instability, with some changes producing up to a 12-fold increase of landslide rate. A relationship between road network and the spatial distribution of landslides has also been observed. These results do confirm the existence of an acceleration of geomorphic processes in the region, and also suggest that climate-related factors play a limited role in the changes observed.