A novel multi-objective medical feature selection compass method for binary classification.

The use of Artificial Intelligence in medical decision support systems has been widely studied. Since a medical decision is frequently the result of a multi-objective optimization problem, a popular challenge combining Artificial Intelligence and Medicine is Multi-Objective Feature Selection (MOFS). This article proposes a novel approach for MOFS applied to medical binary classification. It is built upon a Genetic Algorithm and a 3-Dimensional Compass that aims at guiding the search towards a desired trade-off between: Number of features, Accuracy and Area Under the ROC Curve (AUC). This method, the Genetic Algorithm with multi-objective Compass (GAwC), outperforms all other competitive genetic algorithm-based MOFS approaches on several real-world medical datasets. Moreover, by considering AUC as one of the objectives, GAwC guarantees the classification quality of the solution it provides thus making it a particularly interesting approach for medical problems where both healthy and ill patients should be accurately detected. Finally, GAwC is applied to a real-world medical classification problem and its results are discussed and justified both from a medical point of view and in terms of classification quality.

Identification of new factors associated to walking impairment in patients with vascular-type claudication.

OBJECTIVES: Mechanisms of walking limitation in arterial claudication are incompletely elucidated. We aimed to identify new variables associated to walking limitation in patients with claudication. METHODS: We retrospectively analyzed data of 1120 patients referred for transcutaneous exercise oxygen pressure recordings (TcpO2). The outcome measurement was the absolute walking time on treadmill (3.2 km/h, 10% slope). We used both: linear regression analysis and a non-linear analysis, combining support vector machines and genetic explanatory in 800 patients with the following resting variables: age, gender, body mass index, the presence of diabetes, minimal ankle to brachial index at rest, usual walking speed over 10 m (usual-pace), number of comorbid conditions, active smoking, resting heart rate, pre-test glycaemia and hemoglobin, beta-blocker use, and exercise-derived variables: minimal value of pulse oximetry, resting chest-TcpO2, decrease in chest TcpO2 during exercise, presence of buttock ischemia defined as a decrease from rest of oxygen pressure index ≤15 mmHg. We tested the models over 320 other patients. RESULTS: Independent variables associated to walking time, by decreasing importance in the models, were: age, ankle to brachial index, usual-pace; resting TcpO2, body mass index, smoking, buttock ischemia, heart rate and beta-blockers for the linear regression analysis, and were ankle to brachial index, age, body mass index, usual-pace, decrease in chest TcpO2, smoking, buttock ischemia, glycaemia, heart rate for the non-linear analysis. Testing of models over 320 new patients gave r = 0.509 for linear and 0.575 for non-linear analysis (both p < 0.05). CONCLUSION: Buttock ischemia, heart rate and usual-pace are new variables associated to walking time.

Evaluation of automatic feature detection algorithms in EEG: application to interburst intervals.

In this paper, we present a new method to compare and improve algorithms for feature detection in neonatal EEG. The method is based on the algorithm׳s ability to compute accurate statistics to predict the results of EEG visual analysis. This method is implemented inside a Java software called EEGDiag, as part of an e-health Web portal dedicated to neonatal EEG. EEGDiag encapsulates a component-based implementation of the detection algorithms called analyzers. Each analyzer is defined by a list of modules executed sequentially. As the libraries of modules are intended to be enriched by its users, we developed a process to evaluate the performance of new modules and analyzers using a database of expertized and categorized EEGs. The evaluation is based on the Davies-Bouldin index (DBI) which measures the quality of cluster separation, so that it will ease the building of classifiers on risk categories. For the first application we tested this method on the detection of interburst intervals (IBI) using a database of 394 EEG acquired on premature newborns. We have defined a class of IBI detectors based on a threshold of the standard deviation on contiguous short time windows, inspired by previous work. Then we determine which detector and what threshold values are the best regarding DBI, as well as the robustness of this choice. This method allows us to make counter-intuitive choices, such as removing the 50 Hz filter (power supply) to save time.

Cardiovascular variables do not predict head-up tilt test outcome better than body composition.

Reflex syncope is a well-recognized phenomenon, but the understanding of its underlying pathophysiology remains limited. We hypothesized that patients with a history of syncope and a positive head-up tilt test (HUTT) outcome are in a "not-yet defined" abnormal state even before the head-up position. We performed a 45 min HUTT on 86 patients with a history of syncope. We assessed 19 variables during the supine period before head-up position. Of these variables, 9 were cardiovascular variables (CV) while 8 were body composition variables (BC). The two remaining variables were age and sex. Forty-five patients (41 +/- 15 years, 22 males) have a positive HUTT outcome and 41 a negative one (46 +/- 15 years, 22 males). Statistical tests applied on each of the 19 variables individually did not discriminate patients with a positive and a negative outcome. We used neural networks to screen the sets of variables that allowed for the best predictions of HUTT outcomes. The sensitivity, specificity, positive and negative predictive values (PPV and NPV, respectively) were determined. The best set of predictive values determined from the 19 variables was 76, 81, 78 and 80% for sensitivity, sensibility, PPV, and NPV, respectively. Unexpectedly, the HUTT outcome prediction performed with cardiovascular variables was not better than the prediction performed with body composition variables only. Patients with a positive HUTT outcome are in an abnormal state that can be detected even before the head-up position. Body composition is an important contributor to this abnormal state.

Early prediction of unexplained syncope by support vector machines.

The goal of the present study was to develop and evaluate a new method for the prediction of unexplained syncope occurrences. Diagnosis of syncope is currently based on the reproduction of symptoms in combination with hypotension and bradycardia induced by a 45 min 60-70 degrees head-upright tilt test (HUTT). The main drawback of this widely used test concerns its duration that reaches 55 min if the patient does not faint. Our method is a first step in the avoidance of the HUTT. An electrocardiogram and a transthoracic impedance waveform were recorded for 10 min of supine rest of a HUTT in 128 patients with a history of unexplained recurrent syncope. Seven indices were computed on the transthoracic impedance and its first derivative. The prediction quality of every subset of these variables, mixed with age and sex, has been tested by a support vector machine in a retrospective group of 64 patients (100% of sensitivity and 100% of specificity was reached). The best subset obtained has been evaluated prospectively in a group of 64 patients (94% of sensitivity and 79% of specificity was reached). These results compare very favorably with published results for other unexplained syncope detectors.

Late hemodynamic changes during a negative passive head-up tilt predict the symptomatic outcome to a nitroglycerin sensitized tilt.

BACKGROUND: Sublingual nitroglycerin is advocated to sensitize the passive 70 degrees head-upright tilt test (HUTT) of patients with unexplained syncope. We hypothesized that a detailed analysis of hemodynamic responses recorded during a negative HUTT could predict the outcome to a subsequent nitroglycerin sensitized HUTT (NTG-HUTT). METHODS: Thirty-two patients (46 +/- 3 years, 17 males) with recurrent unexplained syncope but a negative HUTT were included. Heart rate, arterial blood pressure, and central hemodynamics assessed by transthoracic impedance (preejection and rapid left ventricular ejection time, slow ejection time, peak amplitude of first derivative, and cardiac index) were recorded during supine rest and 45 minutes HUTT. Changes from supine rest of the variables were retrospectively compared between patients with a negative (n = 15, NTG-HUTT(-)) and positive (n = 17, NTG-HUTT(+)) outcome to 10 minutes subsequent NTG-HUTT. RESULT: Significant differences between groups were observed during the 15th-20th minutes (Italian protocol) and during the last 5 minutes of passive HUTT (Westminster protocol). The combination of cutoff values, determined by receiver operating curves, on hemodynamic variables changes during the last 5 minutes predicted the outcome to a NTG-HUTT with a sensitivity of 76% and a specificity of 87%. The cutoff values determined during 15th-20th minutes gave an attractive sensitivity (85%) but a too weak specificity (53%) to shorten the 45 minutes passive HUTT at 20 minutes. CONCLUSION: Outcome to a NTG-HUTT can be reliably predicted by selected criteria determined from multiple hemodynamic variables recorded during a passive 70 degrees HUTT.

Changes in the transthoracic impedance signal predict the outcome of a 70 degrees head-up tilt test.

We determined whether early changes in central haemodynamics, as determined by transthoracic impedance, induced by a 70 degrees head-up tilt (HUT) test could predict syncope. Heart rate, arterial blood pressure and central haemodynamics [pre-ejection period and rapid left ventricular ejection time ( T (1)), slow ejection time ( T (2)) and d Z /d t (max) (where Z is thoracic impedance), assessed by the transthoracic impedance technique], were recorded during supine rest and during a 45 min 70 degrees HUT test in 68 patients (40+/-2 years) with a history of unexplained recurrent syncope. We found that 38 patients (42+/-3 years) had a symptomatic outcome to 70 degrees HUT (fainters) and 30 (39+/-2 years) had a negative outcome (non-fainters). When measured between 5 and 10 min of 70 degrees HUT, T (2) had increased significantly only in the fainters, and a change in T (2) of >40 ms from baseline predicted a positive outcome with a sensitivity of 68% and a specificity of 70%. During supine rest prior to 70 degrees HUT, the fainters exhibited a shorter T (2) than non-fainters (183+/-10 compared with 233+/-14 ms; P <0.01), and a T (2) of <199 ms predicted a positive outcome to 70 degrees HUT with a sensitivity of 68% and a specificity of 63%. Incorporation of the changes that occurred from rest to 70 degrees HUT in other haemodynamic variables (heart rate >11 beats/min, systolic pressure <2 mmHg, diastolic pressure <7 mmHg and pulse pressure <-3 mmHg) increased the specificity to 97% and the positive predictive value to 93%. Thus transthoracic impedance could detect differences in central haemodynamics between fainters and non-fainters during supine rest and during the initial period of 70 degrees HUT with a consistent sensitivity and specificity when combined with peripheral haemodynamic variables.