Machine Learning of Bone Marrow Histopathology Identifies Genetic and Clinical Determinants in Patients with MDS.

In myelodysplastic syndrome (MDS) and myeloproliferative neoplasm (MPN), bone marrow (BM) histopathology is assessed to identify dysplastic cellular morphology, cellularity, and blast excess. Yet, other morphologic findings may elude the human eye. We used convolutional neural networks to extract morphologic features from 236 MDS, 87 MDS/MPN, and 11 control BM biopsies. These features predicted genetic and cytogenetic aberrations, prognosis, age, and gender in multivariate regression models. Highest prediction accuracy was found for TET2 [area under the receiver operating curve (AUROC) = 0.94] and spliceosome mutations (0.89) and chromosome 7 monosomy (0.89). Mutation prediction probability correlated with variant allele frequency and number of affected genes per pathway, demonstrating the algorithms' ability to identify relevant morphologic patterns. By converting regression models to texture and cellular composition, we reproduced the classical del(5q) MDS morphology consisting of hypolobulated megakaryocytes. In summary, this study highlights the potential of linking deep BM histopathology with genetics and clinical variables. SIGNIFICANCE: Histopathology is elementary in the diagnostics of patients with MDS, but its high-dimensional data are underused. By elucidating the association of morphologic features with clinical variables and molecular genetics, this study highlights the vast potential of convolutional neural networks in understanding MDS pathology and how genetics is reflected in BM morphology. See related commentary by Elemento, p. 195.

Characteristics of young men with clustering risk factors for coronary heart disease: The cardiovascular risk in young finns study.

The majority of coronary heart disease (CHD) arises from a population with only moderately elevated risk factor levels. This study addressed the issue of whether clustering of moderately adverse levels of serum low density lipoprotein cholesterol (LDL-C), serum high density lipoprotein cholesterol (HDL-C), and diastolic blood pressure (DBP) was detectable in children and young adults. Significant clustering of these factors was observed in young men (ages 18-24 years) (Observed/Expected [O/E] ratio = 1.5, p = 0.014), whereas in young women or children no clustering greater than due to chance was found. In males, clustering tended to increase with age. Compared to young women, young men had a higher relative intake of dietary fat, and smoked and used alcohol more often. Compared to men without risk factors, the men at risk were significantly more obese, consumed more dietary fat, and smoked more often. The highest degree of clustering was seen in the highest quartile for the subscapular skinfold thickness (O/E ratio = 2.4). In conclusion, high LDL-C, low HDL-C, and high DBP cluster in young adult males. The clear sex difference observed in clustering may be one of the causes for the susceptibility of adult men to CHD. Am. J. Hum. Biol. 9:565-571, 1997. © 1997 Wiley-Liss, Inc.