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.

Thrombopoietin receptor-independent stimulation of hematopoietic stem cells by eltrombopag.

Eltrombopag (EP), a small-molecule thrombopoietin receptor (TPO-R) agonist and potent intracellular iron chelator, has shown remarkable efficacy in stimulating sustained multilineage hematopoiesis in patients with bone marrow failure syndromes, suggesting an effect at the most immature hematopoietic stem and multipotent progenitor level. Although the functional and molecular effects of EP on megakaryopoiesis have been studied in the past, mechanistic insights into its effects on the earliest stages of hematopoiesis have been limited. We investigated the effects of EP treatment on hematopoietic stem cell (HSC) function using purified primary HSCs in separation-of-function mouse models, including a TPO-R-deficient strain, and stem cells isolated from patients undergoing TPO-R agonist treatment. Our mechanistic studies showed a stimulatory effect on stem cell self-renewal independently of TPO-R. Human and mouse HSCs responded to acute EP treatment with metabolic and gene expression alterations consistent with a reduction of intracellular labile iron pools that are essential for stem cell maintenance. Iron preloading prevented the stem cell stimulatory effects of EP. Moreover, comparative analysis of stem cells in the bone marrow of patients receiving EP showed a marked increase in the number of functional stem cells compared to patients undergoing therapy with romiplostim, another TPO-R agonist lacking an iron-chelating ability. Together, our study demonstrates that EP stimulates hematopoiesis at the stem cell level through iron chelation-mediated molecular reprogramming and indicates that labile iron pool-regulated pathways can modulate HSC function.

Intelligent composting assisted by a wireless sensing network.

Monitoring of the moisture and temperature of composting process is a key factor to obtain a quality product beyond the quality of raw materials. Current methodologies for monitoring these two parameters are time consuming for workers, sometimes not sufficiently reliable to help decision-making and thus are ignored in some cases. This article describes an advance on monitoring of composting process through a Wireless Sensor Network (WSN) that allows measurement of temperature and moisture in real time in multiple points of the composting material, the Compo-ball system. To implement such measurement capabilities on-line, a WSN composed of multiple sensor nodes was designed and implemented to provide the staff with an efficient monitoring composting management tool. After framing the problem, the objectives and characteristics of the WSN are briefly discussed and a short description of the hardware and software of the network's components are presented. Presentation and discussion of practical issues and results obtained with the WSN during a demonstration stage that took place in several composting sites concludes the paper.