Concept: Decision tree
Inflammatory bowel disease (IBD) and alimentary lymphoma (ALA) are common gastrointestinal diseases in cats. The very similar clinical signs and histopathologic features of these diseases make the distinction between them diagnostically challenging. We tested the use of supervised machine-learning algorithms to differentiate between the 2 diseases using data generated from noninvasive diagnostic tests. Three prediction models were developed using 3 machine-learning algorithms: naive Bayes, decision trees, and artificial neural networks. The models were trained and tested on data from complete blood count (CBC) and serum chemistry (SC) results for the following 3 groups of client-owned cats: normal, inflammatory bowel disease (IBD), or alimentary lymphoma (ALA). Naive Bayes and artificial neural networks achieved higher classification accuracy (sensitivities of 70.8% and 69.2%, respectively) than the decision tree algorithm (63%, p < 0.0001). The areas under the receiver-operating characteristic curve for classifying cases into the 3 categories was 83% by naive Bayes, 79% by decision tree, and 82% by artificial neural networks. Prediction models using machine learning provided a method for distinguishing between ALA-IBD, ALA-normal, and IBD-normal. The naive Bayes and artificial neural networks classifiers used 10 and 4 of the CBC and SC variables, respectively, to outperform the C4.5 decision tree, which used 5 CBC and SC variables in classifying cats into the 3 classes. These models can provide another noninvasive diagnostic tool to assist clinicians with differentiating between IBD and ALA, and between diseased and nondiseased cats.
While associations between specific risk factors and subsequent suicidal thoughts or behaviours have been widely examined, there is limited understanding of the interplay between risk factors in the development of suicide risk. This study used a decision tree approach to develop individual models of suicide risk and identify the risk factors for suicidality that are important for different subpopulations.
Purpose of the Study: A novel logistic regression tree-based method was applied to identify fall risk factors and possible interaction effects of those risk factors.
Epilepsy is a global disease with considerable incidence due to recurrent unprovoked seizures. These seizures can be noninvasively diagnosed using electroencephalogram (EEG), a measure of neuronal electrical activity in brain recorded along scalp. EEG is highly nonlinear, nonstationary and non-Gaussian in nature. Nonlinear adaptive models such as empirical mode decomposition (EMD) provide intuitive understanding of information present in these signals. In this study a novel methodology is proposed to automatically classify EEG of normal, inter-ictal and ictal subjects using EMD decomposition. EEG decomposition using EMD yields few intrinsic mode functions (IMF), which are amplitude and frequency modulated (AM and FM) waves. Hilbert transform of these IMF provides AM and FM frequencies. Features such as spectral peaks, spectral entropy and spectral energy in each IMF are extracted and fed to decision tree classifier for automated diagnosis. In this work, we have compared the performance of classification using two types of decision trees (i) classification and regression tree (CART) and (ii) C4.5. We have obtained the highest average accuracy of 95.33%, average sensitivity of 98%, and average specificity of 97% using C4.5 decision tree classifier. The developed methodology is ready for clinical validation on large databases and can be deployed for mass screening.
OBJECTIVE: Individuals living with HIV face challenging employment decisions that have personal, financial, and health impacts. The decision to stay or to leave the work force is much more complicated for an individual with HIV because the financial choices related to potential health benefits are not clearly understood. To assist in the decision-making process for an individual with HIV, we propose to develop a decision model that compares the potential costs and benefits of staying in or leaving the work force. PARTICIPANTS: A hypothetical cohort of HIV-infected individuals was simulated in our decision model. Characteristics of these individuals over a one-year period were extracted from the medical literature and publicly available national surveys. Men and women between the ages of 18 and 59 were included in our simulated cohort. METHODS: A decision tree model was created to estimate the financial impact of an individual’s decision on employment. The outcomes were presented as the cost-savings associated with the following employment statuses over a one-year period: 1) staying full-time, 2) switching from full-to part-time, 3) transitioning from full-time to unemployment, and 4) staying unemployed. CD4 T cell counts and employment statuses were stratified by earned income. Employment probabilities were calculated from national databases on employment trends in the United States. Sensitivity analyses were conducted to test the robustness of the effects of the variables on the outcomes. RESULTS: Overall, the decision outcome that resulted in the least financial loss for individuals with HIV was to remain at work. For an individual with CD4 T cell count > 350, the cost difference between staying employed full-time and switching from full-time to part-time status was a maximum of $2,970. For an individual with a CD4 T cell count between 200 and 350, the cost difference was as low as $126 and as great as $2,492. For an individual with a CD4 T cell count < 200, the minimum cost difference was $375 and the maximum cost difference was $2,253. CONCLUSIONS: Based on our simulated model, we recommend an individual with CD4 T cell count > 350 to stay employed full-time because it resulted in the least financial loss. On the other hand, for an individual with a CD4 T cell > 350, the financial cost loss was much more variable. Our model provides an objective decision-making guide for individuals with HIV to weigh the costs and benefits of employment decisions.
Ultrasound guided aspiration of ovarian endometrioma had been tried as an alternative therapeutic modality in patients whose desire to avoid surgery or surgical approach is contraindicated since 1991. Cyst puncture can reduce tumor volume and destruct the cyst wall, alleviate sticking circumstances and enhance the chance of recovery. But simple aspiration without other treatments results in high recurrence rate (28.5 % to 100 %). In order to reduce recurrence after aspiration, ultrasound-guided aspiration with instillation of tetracycline, methotrexate, and recombinant interleukin-2 has been combined and proven to be effective with the recurrence rates of 46.9 %, 18.1 %, and 40 % respectively. Noma et al. (2001) reported that conduct of ethanol instillation for more than 10 min particularly for a case with a single endometrial cyst is considered most effective from the standpoint of recurrence (14.9 %). Our goal is to analyze patients with recurrent pelvic cyst who underwent surgical intervention. The research data are based on clinical diagnosis, symptoms and medical intervention classification, and the cyst numbers are defined as forecast project target. The decision tree, methodology of data mining technology, is used to find the meaningful characteristic as well as each other mutually connection. The experimental result can help the clinical faculty doctors to better diagnose and provide treatment reference for future patients.
Unintended pregnancy is reportedly higher in active duty women; therefore, we sought to estimate the potential impact of the levonorgestrel-containing intrauterine system (LNG-IUS) could have on unintended pregnancy in active duty women. A decision tree model with sensitivity analysis was used to estimate the number of unintentional pregnancies in active duty women which could be prevented. A secondary cost analysis was performed to analyze the direct cost savings to the U.S. Government. The total number of Armed Services members is estimated to be over 1.3 million, with an estimated 208,146 being women. Assuming an age-standardized unintended pregnancy rate of 78 per 1,000 women, 16,235 unintended pregnancies occur each year. Using a combined LNG-IUS failure and expulsion rate of 2.2%, a decrease of 794, 1588, and 3970 unintended pregnancies was estimated to occur with 5%, 10% and 25% usage, respectively. Annual cost savings from LNG-IUS use range from $3,387,107 to $47,352,295 with 5% to 25% intrauterine device usage. One-way sensitivity analysis demonstrated LNG-IUS to be cost-effective when the cost associated with pregnancy and delivery exceeded $11,000. Use of LNG-IUS could result in significant reductions in unintended pregnancy among active duty women, resulting in substantial cost savings to the government health care system.
We compared the expected medical costs of empirical and preemptive treatment strategies for invasive fungal infection in neutropenic patients with hematological diseases. Based on the results of two clinical trials with different backgrounds reported by Oshima et al. [J Antimicrob Chemother 60(2):350-355; Oshima study] and Cordonnier et al. [Clin Infect Dis 48(8):1042-1051; PREVERT study], we developed a decision tree model that represented the outcomes of empirical and preemptive treatment strategies, and estimated the expected medical costs of medications and examinations in the two strategies. We assumed that micafungin was started in the empirical group at 5 days after fever had developed, while voriconazole was started in the preemptive group only when certain criteria, such as positive test results of imaging studies and/or serum markers, were fulfilled. When we used an incidence of positive test results of 6.7 % based on the Oshima study, the expected medical costs of the empirical and preemptive groups were 288,198 and 150,280 yen, respectively. Even in the case of the PREVERT study, in which the incidence of positive test results was 32.9 %, the expected medical costs in the empirical and preemptive groups were 291,871 and 284,944 yen, respectively. A sensitivity analysis indicated that the expected medical costs in the preemptive group would exceed those in the empirical group when the incidence of positive test results in the former was over 34.4 %. These results suggest that a preemptive treatment strategy can be expected to reduce medical costs compared with empirical therapy in most clinical settings.
Escalating drug prices have catalysed the generation of numerous “value frameworks” with the aim of informing payers, clinicians and patients on the assessment and appraisal process of new medicines for the purpose of coverage and treatment selection decisions. Although this is an important step towards a more inclusive Value Based Assessment (VBA) approach, aspects of these frameworks are based on weak methodologies and could potentially result in misleading recommendations or decisions. In this paper, a Multiple Criteria Decision Analysis (MCDA) methodological process, based on Multi Attribute Value Theory (MAVT), is adopted for building a multi-criteria evaluation model. A five-stage model-building process is followed, using a top-down “value-focused thinking” approach, involving literature reviews and expert consultations. A generic value tree is structured capturing decision-makers' concerns for assessing the value of new medicines in the context of Health Technology Assessment (HTA) and in alignment with decision theory. The resulting value tree (Advance Value Tree) consists of three levels of criteria (top level criteria clusters, mid-level criteria, bottom level sub-criteria or attributes) relating to five key domains that can be explicitly measured and assessed: (a) burden of disease, (b) therapeutic impact, © safety profile (d) innovation level and (e) socioeconomic impact. A number of MAVT modelling techniques are introduced for operationalising (i.e. estimating) the model, for scoring the alternative treatment options, assigning relative weights of importance to the criteria, and combining scores and weights. Overall, the combination of these MCDA modelling techniques for the elicitation and construction of value preferences across the generic value tree provides a new value framework (Advance Value Framework) enabling the comprehensive measurement of value in a structured and transparent way. Given its flexibility to meet diverse requirements and become readily adaptable across different settings, the Advance Value Framework could be offered as a decision-support tool for evaluators and payers to aid coverage and reimbursement of new medicines.
Computer simulation of genomic data has become increasingly popular for assessing and validating biological models or for gaining an understanding of specific data sets. Several computational tools for the simulation of next-generation sequencing (NGS) data have been developed in recent years, which could be used to compare existing and new NGS analytical pipelines. Here we review 23 of these tools, highlighting their distinct functionality, requirements and potential applications. We also provide a decision tree for the informed selection of an appropriate NGS simulation tool for the specific question at hand.