RESUMO
Decision-making related to health is complex. Machine learning (ML) and patient generated data can identify patterns and insights at the individual level, where human cognition falls short, but not all ML-generated information is of equal utility for making health-related decisions. We develop and apply attributable components analysis (ACA), a method inspired by optimal transport theory, to type 2 diabetes self-monitoring data to identify patterns of association between nutrition and blood glucose control. In comparison with linear regression, we found that ACA offers a number of characteristics that make it promising for use in decision support applications. For example, ACA was able to identify non-linear relationships, was more robust to outliers, and offered broader and more expressive uncertainty estimates. In addition, our results highlight a tradeoff between model accuracy and interpretability, and we discuss implications for ML-driven decision support systems.
Assuntos
Diabetes Mellitus Tipo 2 , Diabetes Mellitus Tipo 2/diagnóstico , Diabetes Mellitus Tipo 2/terapia , Humanos , Aprendizado de MáquinaRESUMO
Self-tracking can help personalize self-management interventions for chronic conditions like type 2 diabetes (T2D), but reflecting on personal data requires motivation and literacy. Machine learning (ML) methods can identify patterns, but a key challenge is making actionable suggestions based on personal health data. We introduce GlucoGoalie, which combines ML with an expert system to translate ML output into personalized nutrition goal suggestions for individuals with T2D. In a controlled experiment, participants with T2D found that goal suggestions were understandable and actionable. A 4-week in-the-wild deployment study showed that receiving goal suggestions augmented participants' self-discovery, choosing goals highlighted the multifaceted nature of personal preferences, and the experience of following goals demonstrated the importance of feedback and context. However, we identified tensions between abstract goals and concrete eating experiences and found static text too ambiguous for complex concepts. We discuss implications for ML-based interventions and the need for systems that offer more interactivity, feedback, and negotiation.
RESUMO
Motor output varies along the rostro-caudal axis of the tetrapod spinal cord. At limb levels, â¼60 motor pools control the alternation of flexor and extensor muscles about each joint, whereas at thoracic levels as few as 10 motor pools supply muscle groups that support posture, inspiration, and expiration. Whether such differences in motor neuron identity and muscle number are associated with segmental distinctions in interneuron diversity has not been resolved. We show that select combinations of nineteen transcription factors that specify lumbar V1 inhibitory interneurons generate subpopulations enriched at limb and thoracic levels. Specification of limb and thoracic V1 interneurons involves the Hox gene Hoxc9 independently of motor neurons. Thus, early Hox patterning of the spinal cord determines the identity of V1 interneurons and motor neurons. These studies reveal a developmental program of V1 interneuron diversity, providing insight into the organization of inhibitory interneurons associated with differential motor output.
Assuntos
Genes Homeobox , Medula Espinal/citologia , Animais , Teorema de Bayes , Membro Anterior/embriologia , Membro Anterior/inervação , Perfilação da Expressão Gênica , Membro Posterior/embriologia , Membro Posterior/inervação , Proteínas de Homeodomínio/fisiologia , Interneurônios/fisiologia , Região Lombossacral , Camundongos , Camundongos Knockout , Neurônios Motores/fisiologia , Proteínas do Tecido Nervoso/fisiologia , Medula Espinal/embriologia , Tórax , Fatores de Transcrição/fisiologiaRESUMO
Background. Cardiac allograft vasculopathy (CAV) is a major cause of graft loss and death after heart transplantation. Currently, no diagnostic methods are available during the early post-transplant period to accurately identify patients at risk of CAV. We hypothesized that PBMC gene expression profiles (GEP) can identify patients at risk of CAV. Methods. We retrospectively analyzed a limited set of whole-genome PBMC microarrays from 10 post-transplant patients who did (n = 3) or did not (n = 7) develop advanced grade CAV during their long-term follow-up. We used significance analysis of microarrays to identify differentially expressed genes and High-Throughput GoMiner to assess gene ontology (GO) categories. We corroborated our findings by retrospective analysis of PBMC real-time PCR data from 33 patients. Results. Over 300 genes were differentially expressed (FDR < 5%), and 18 GO-categories including "macrophage activation", "Interleukin-6 pathway", "NF-KappaB cascade", and "response to virus" were enriched by these genes (FDR < 5%). Out of 8 transcripts available for RT-PCR analysis, we confirmed 6 transcripts (75.0%) including FPRL1, S100A9, CXCL10, PRO1073, and MMP9 (P < .05). Conclusion. Our pilot data suggest that GEP of PBMC may become a valuable tool in the evaluation of patients at risk of CAV. Larger prospectively designed studies are needed to corroborate our hypothesis.
RESUMO
The high-frequency limit of the Garrett and Munk spectrum of internal waves in the ocean and the observed deviations from it are shown to form a pattern consistent with the predictions of wave turbulence theory. In particular, the high-frequency limit of the Garrett and Munk spectrum constitutes an exact steady-state solution of the corresponding kinetic equation.