Performance and usability testing of an automated tool for detection of peripheral artery disease using electronic health records.

Peripheral artery disease (PAD) is a common cardiovascular disorder that is frequently underdiagnosed, which can lead to poorer outcomes due to lower rates of medical optimization. We aimed to develop an automated tool to identify undiagnosed PAD and evaluate physician acceptance of a dashboard representation of risk assessment. Data were derived from electronic health records (EHR). We developed and compared traditional risk score models to novel machine learning models. For usability testing, primary and specialty care physicians were recruited and interviewed until thematic saturation. Data from 3168 patients with PAD and 16,863 controls were utilized. Results showed a deep learning model that utilized time engineered features outperformed random forest and traditional logistic regression models (average AUCs 0.96, 0.91 and 0.81, respectively), P < 0.0001. Of interviewed physicians, 75% were receptive to an EHR-based automated PAD model. Feedback emphasized workflow optimization, including integrating risk assessments directly into the EHR, using dashboard designs that minimize clicks, and providing risk assessments for clinically complex patients. In conclusion, we demonstrate that EHR-based machine learning models can accurately detect risk of PAD and that physicians are receptive to automated risk detection for PAD. Future research aims to prospectively validate model performance and impact on patient outcomes.

Clonal Expansion of Stem/Progenitor Cells in Cancer, Fibrotic Diseases, and Atherosclerosis, and CD47 Protection of Pathogenic Cells.

We proposed and demonstrated that myelogenous leukemia has a preleukemic phase. In the premalignant phase, normal hematopoietic stem cells (HSCs) gradually accumulate mutations leading to HSC clonal expansion, resulting in the emergence of leukemic stem cells (LSCs). Here, we show that preleukemic HSCs are the basis of clonal hematopoiesis, as well as late-onset blood diseases (chronic-phase chronic myeloid leukemia, myeloproliferative neoplasms, and myelodysplastic disease). The clones at some point each trigger surface expression of "eat me" signals for macrophages, and in the clones and their LSC progeny, this is countered by upregulation of "don't eat me" signals for macrophages such as CD47,opening the possibility of CD47-based therapies. We include evidence that similar processes result in fibroblast expansion in a variety of fibrotic diseases, and arterial smooth muscle clonal expansion is a basis of atherosclerosis, including upregulation of both "eat me" and "don't eat me" molecules on the pathogenic cells.

Cloning of a unique lipase from endothelial cells extends the lipase gene family.

A new lipoprotein lipase-like gene has been cloned from endothelial cells through a subtraction methodology aimed at characterizing genes that are expressed with in vitro differentiation of this cell type. The conceptual endothelial cell-derived lipase protein contains 500 amino acids, including an 18-amino acid hydrophobic signal sequence, and is 44% identical to lipoprotein lipase and 41% identical to hepatic lipase. Comparison of primary sequence to that of lipoprotein and hepatic lipase reveals conservation of the serine, aspartic acid, and histidine catalytic residues as well as the 10 cysteine residues involved in disulfide bond formation. Expression was identified in cultured human umbilical vein endothelial cells, human coronary artery endothelial cells, and murine endothelial-like yolk sac cells by Northern blot. In addition, Northern blot and in situ hybridization analysis revealed expression of the endothelial-derived lipase in placenta, liver, lung, ovary, thyroid gland, and testis. A c-Myc-tagged protein secreted from transfected COS7 cells had phospholipase A1 activity but no triglyceride lipase activity. Its tissue-restricted pattern of expression and its ability to be expressed by endothelial cells, suggests that endothelial cell-derived lipase may have unique functions in lipoprotein metabolism and in vascular disease.

Tetracycline-regulatable expression vectors tightly regulate in vitro gene expression of secreted proteins.

The regulation of gene expression by the tetracycline system has attracted a high level of interest in the recent past. However, expression of secreted proteins has not been evaluated precisely. In this study, we constructed two versions of a one-plasmid system containing the elements necessary for the regulation of gene expression. The regulatable elements and the selectable marker (Neor) were set up in two different configurations, pTRIN31 and pTRIN76. With these two regulatable versions, the levels of protein expression after transfection into the NIH/3T3 cell line were measured by insertion of three different genes encoding the secreted proteins (hGH, ApoE3, hGM-CSF). The maximum levels of gene expression obtained with the pTRIN76-derived plasmids were 100ng/24h/106 cells for hGH, 427ng/24h/106 cells for ApoE3 and 108ng/24h/106 cells for hGM-CSF. For the pTRIN31-derived plasmids the maximum levels were 2.7ng/24h/106 cells for hGH and 47ng/24h/106 for ApoE3. Both plasmids give rise to an expression of the transfected gene that can be tightly regulated by three different molecules: tetracycline, minocycline and doxycycline. The levels of the secreted proteins are below the detectable level when the reporter genes are repressed. This repression is reversible within 48h after the regulator has been removed from the medium.