Human-in-the-Loop-A Deep Learning Strategy in Combination with a Patient-Specific Gaussian Mixture Model Leads to the Fast Characterization of Volumetric Ground-Glass Opacity and Consolidation in the Computed Tomography Scans of COVID-19 Patients.

Background/Objectives: The accurate quantification of ground-glass opacities (GGOs) and consolidation volumes has prognostic value in COVID-19 patients. Nevertheless, the accurate manual quantification of the corresponding volumes remains a time-consuming task. Deep learning (DL) has demonstrated good performance in the segmentation of normal lung parenchyma and COVID-19 pneumonia. We introduce a Human-in-the-Loop (HITL) strategy for the segmentation of normal lung parenchyma and COVID-19 pneumonia that is both time efficient and quality effective. Furthermore, we propose a Gaussian Mixture Model (GMM) to classify GGO and consolidation based on a probabilistic characterization and case-sensitive thresholds. Methods: A total of 65 Computed Tomography (CT) scans from 64 patients, acquired between March 2020 and June 2021, were randomly selected. We pretrained a 3D-UNet with an international dataset and implemented a HITL strategy to refine the local dataset with delineations by teams of medical interns, radiology residents, and radiologists. Following each HITL cycle, 3D-UNet was re-trained until the Dice Similarity Coefficients (DSCs) reached the quality criteria set by radiologists (DSC = 0.95/0.8 for the normal lung parenchyma/COVID-19 pneumonia). For the probabilistic characterization, a Gaussian Mixture Model (GMM) was fitted to the Hounsfield Units (HUs) of voxels from the CT scans of patients with COVID-19 pneumonia on the assumption that two distinct populations were superimposed: one for GGO and one for consolidation. Results: Manual delineation of the normal lung parenchyma and COVID-19 pneumonia was performed by seven teams on 65 CT scans from 64 patients (56 ± 16 years old (µ ± σ), 46 males, 62 with reported symptoms). Automated lung/COVID-19 pneumonia segmentation with a DSC > 0.96/0.81 was achieved after three HITL cycles. The HITL strategy improved the DSC by 0.2 and 0.5 for the normal lung parenchyma and COVID-19 pneumonia segmentation, respectively. The distribution of the patient-specific thresholds derived from the GMM yielded a mean of -528.4 ± 99.5 HU (µ ± σ), which is below most of the reported fixed HU thresholds. Conclusions: The HITL strategy allowed for fast and effective annotations, thereby enhancing the quality of segmentation for a local CT dataset. Probabilistic characterization of COVID-19 pneumonia by the GMM enabled patient-specific segmentation of GGO and consolidation. The combination of both approaches is essential to gain confidence in DL approaches in our local environment. The patient-specific probabilistic approach, when combined with the automatic quantification of COVID-19 imaging findings, enhances the understanding of GGO and consolidation during the course of the disease, with the potential to improve the accuracy of clinical predictions.

Quantifying massively parallel microbial growth with spatially mediated interactions.

Quantitative understanding of microbial growth is an essential prerequisite for successful control of pathogens as well as various biotechnology applications. Even though the growth of cell populations has been extensively studied, microbial growth remains poorly characterised at the spatial level. Indeed, even isogenic populations growing at different locations on solid growth medium typically show significant location-dependent variability in growth. Here we show that this variability can be attributed to the initial physiological states of the populations, the interplay between populations interacting with their local environment and the diffusion of nutrients and energy sources coupling the environments. We further show how the causes of this variability change throughout the growth of a population. We use a dual approach, first applying machine learning regression models to discover that location dominates growth variability at specific times, and, in parallel, developing explicit population growth models to describe this spatial effect. In particular, treating nutrient and energy source concentration as a latent variable allows us to develop a mechanistic resource consumer model that captures growth variability across the shared environment. As a consequence, we are able to determine intrinsic growth parameters for each local population, removing confounders common to location-dependent variability in growth. Importantly, our explicit low-parametric model for the environment paves the way for massively parallel experimentation with configurable spatial niches for testing specific eco-evolutionary hypotheses.

Analysis of Tumor-Infiltrating T-Cell Transcriptomes Reveal a Unique Genetic Signature across Different Types of Cancer.

CD8+ and CD4+ T-cells play a key role in cellular immune responses against cancer by cytotoxic responses and effector lineages differentiation, respectively. These subsets have been found in different types of cancer; however, it is unclear whether tumor-infiltrating T-cell subsets exhibit similar transcriptome profiling across different types of cancer in comparison with healthy tissue-resident T-cells. Thus, we analyzed the single cell transcriptome of five tumor-infiltrating CD4-T, CD8-T and Treg cells obtained from different types of cancer to identify specific pathways for each subset in malignant environments. An in silico analysis was performed from single-cell RNA-sequencing data available in public repositories (Gene Expression Omnibus) including breast cancer, melanoma, colorectal cancer, lung cancer and head and neck cancer. After dimensionality reduction, clustering and selection of the different subpopulations from malignant and nonmalignant datasets, common genes across different types of cancer were identified and compared to nonmalignant genes for each T-cell subset to identify specific pathways. Exclusive pathways in CD4+ cells, CD8+ cells and Tregs, and common pathways for the tumor-infiltrating T-cell subsets were identified. Finally, the identified pathways were compared with RNAseq and proteomic data obtained from T-cell subsets cultured under malignant environments and we observed that cytokine signaling, especially Th2-type cytokine, was the top overrepresented pathway in Tregs from malignant samples.

Secretome from Human Mesenchymal Stem Cells-Derived Endothelial Cells Promotes Wound Healing in a Type-2 Diabetes Mouse Model.

Tissue regeneration is often impaired in patients with metabolic disorders such as diabetes mellitus and obesity, exhibiting reduced wound repair and limited regeneration capacity. We and others have demonstrated that wound healing under normal metabolic conditions is potentiated by the secretome of human endothelial cell-differentiated mesenchymal stem cells (hMSC-EC). However, it is unknown whether this effect is sustained under hyperglycemic conditions. In this study, the wound healing effect of secretomes from undifferentiated human mesenchymal stem cells (hMSC) and hMSC-EC in a type-2 diabetes mouse model was analyzed. hMSC were isolated from human Wharton's jelly and differentiated into hMSC-EC. hMSC and hMSC-EC secretomes were analyzed and their wound healing capacity in C57Bl/6J mice fed with control (CD) or high fat diet (HFD) was evaluated. Our results showed that hMSC-EC secretome enhanced endothelial cell proliferation and wound healing in vivo when compared with hMSC secretome. Five soluble proteins (angiopoietin-1, angiopoietin-2, Factor de crecimiento fibroblástico, Matrix metallopeptidase 9, and Vascular Endothelial Growth Factor) were enriched in hMSC-EC secretome in comparison to hMSC secretome. Thus, the five recombinant proteins were mixed, and their pro-healing property was evaluated in vitro and in vivo. Functional analysis demonstrated that a cocktail of these proteins enhanced the wound healing process similar to hMSC-EC secretome in HFD mice. Overall, our results show that hMSC-EC secretome or a combination of specific proteins enriched in the hMSC-EC secretome enhanced wound healing process under hyperglycemic conditions.