Beyond the routine CBC: machine learning and statistical analyses identify research CBC parameter associations with myelodysplastic syndromes and specific underlying pathogenic variants.

AIMS: Given the time, expense and clinical expertise required for a myelodysplastic syndrome (MDS) diagnosis, there is a clear need for a cost-effective screening laboratory test that can rapidly and accurately distinguish patients with cytopenias related to MDS from other causes. METHODS: We measured conventional and research use only complete blood cell (CBC) parameters using the Sysmex XN-series haematology analyser in 102 MDS patients (70 patients with active MDS and 32 patients in remission), 43 patients with cytopenia without morphological evidence of MDS and 484 age-adjusted controls. A variety of algorithms, including random forest machine learning, were used to construct parameter-based models to predict the presence of MDS using both CBC and molecular data or CBC data alone and correlated individual pathogenic variants/genetic pathways with CBC parameters changes. RESULTS: Using the CBC parameters alone, our predictive model for active MDS showed a 0.86 receiver operating characteristic curve (ROC)/area under the ROC curve (AUC), with 0.87 sensitivity and 0.72 specificity; with the addition of the molecular and demographic status, the ROC/AUC improved to 0.93, sensitivity to 0.89 and specificity to 0.84. The most discriminatory MDS parameters were reflective of dysplastic neutrophil morphology, red cell count fragmentation and degree of platelet immaturity. Specific patterns of parameters were associated with individual gene pathogenic variants or affected pathways. CONCLUSIONS: CBC research parameters can be used as an adjunct to the haematological workup of cytopenia(s) to help screen for patients with high likelihood of MDS.

Modeling fMRI data generated by overlapping cognitive processes with unknown onsets using Hidden Process Models.

We present a new method for modeling fMRI time series data called Hidden Process Models (HPMs). Like several earlier models for fMRI analysis, Hidden Process Models assume that the observed data is generated by a sequence of underlying mental processes that may be triggered by stimuli. HPMs go beyond these earlier models by allowing for processes whose timing may be unknown, and that might not be directly tied to specific stimuli. HPMs provide a principled, probabilistic framework for simultaneously learning the contribution of each process to the observed data, as well as the timing and identities of each instantiated process. They also provide a framework for evaluating and selecting among competing models that assume different numbers and types of underlying mental processes. We describe the HPM framework and its learning and inference algorithms, and present experimental results demonstrating its use on simulated and real fMRI data. Our experiments compare several models of the data using cross-validated data log-likelihood in an fMRI study involving overlapping mental processes whose timings are not fully known.

Block-suffix shifting: fast, simultaneous medical concept set identification in large medical record corpora.

Owing to new advances in computer hardware, large text databases have become more prevalent than ever.Automatically mining information from these databases proves to be a challenge due to slow pattern/string matching techniques. In this paper we present a new, fast multi-string pattern matching method based on the well known Aho-Chorasick algorithm. Advantages of our algorithm include:the ability to exploit the natural structure of text, the ability to perform significant character shifting, avoiding backtracking jumps that are not useful, efficiency in terms of matching time and avoiding the typical "sub-string" false positive errors.Our algorithm is applicable to many fields with free text, such as the health care domain and the scientific document field. In this paper, we apply the BSS algorithm to health care data and mine hundreds of thousands of medical concepts from a large Electronic Medical Record (EMR) corpora simultaneously and efficiently. Experimental results show the superiority of our algorithm when compared with the top of the line multi-string matching algorithms.

Modeling the fMRI signal via Hierarchical Clustered Hidden Process Models.

Machine Learning techniques have been used quite widely for the task of predicting cognitive processes from fMRI data. However, these models do not describe well the fMRI signal when it is generated by multiple cognitive processes that are simultaneously active. In this paper we consider the problem of accurately modeling the fMRI signal of a human subject who is performing a task involving multiple concurrent cognitive processes. We present a Hierarchical Clustering extension of Hidden Process Models which, by taking advantage of automatically discovered similarities in the activation among neighboring voxels, achieves significantly better performance than standard generative models in terms of Average Log Likelihood.