A new scheme for the automatic assessment of Alzheimer's disease on a fine motor task with Transfer Learning.

We present a new scheme for Alzheimer's Disease (AD) automatic assessment, based on Archimedes spiral, drawn on a digitizing tablet. We propose to enrich spiral images generated from the raw sequence of pen coordinates with dynamic information (pressure, altitude, velocity) represented with a semi-global encoding in RGB images. By exploiting Transfer Learning, such hybrid images are given as input to a deep network for an automatic high-level feature extraction. Experiments on 30 AD patients and 45 Healthy Controls (HC) showed that the hybrid representations allow a considerable improvement of classification performance, compared to those obtained on raw spiral images. We reach, with SVM classifiers, an accuracy of 79% with pressure, 76% with velocity, and 70.5% with altitude. The analysis with PCA of internal features of the deep network, showed that dynamic information included in images explain a much higher amount of variance compared to raw images. Moreover, our study demonstrates the need for a semi-global description of dynamic parameters, for a better discrimination of AD and HC classes. This description allows uncovering specific trends on the dynamics for both classes. Finally, combining the decisions of the three SVMs leads to 81.5% of accuracy.Clinical Relevance- This work proposes a decision-aid tool for detecting AD at an early stage, based on a non-invasive simple graphic task, executed on a Wacom digitizer. This task can be considered in the battery of usual clinical tests.

Combined effect of alkali pretreatment and sodium chloride addition on the olive fermentation process.

Green olives of the Tunisian variety "Meski" were treated according to a Spanish-style green olive preservation process by using an alkaline treatment (1.5, 2 and 2.5% (w/v) NaOH) to eliminate bitterness, combined with different brine concentrations (6, 9 and 12% (w/v) NaCl). A spontaneous fermentation by the environmental microflora took place. Results showed that 2% NaOH solution and 9% sodium chloride brine was an optimal combination inducing the best growth of Lactobacillus species (10(8) CFU/ml) and acidity of 0.726 g lactic acid/100 ml brine. In all trials and independently of the treatment, Lb. plantarum was the most dominant strain of Lactobacillus. Moreover, pretreatment with lye and lactic fermentation of olives contributed to coliform elimination.

Work-exhaustion time relationships and the critical power concept. A critical review.

The present review is focused on the physiological meanings of the critical power concept proposed by Scherrer in 1954 and its applications to general exercises such as running, cycling and swimming. Since the first studies on the critical power of local exercises, many studies have found that critical power is correlated with indices which are related to aerobic endurance such as maximal oxygen uptake, ventilatory threshold, OBLA or maximal lactate steady state. In fact, the relationship between exhaustion time t(lim) and the Work Wlim (or Distance Dlim) performed at exhaustion is not exactly linear and, consequently, the power-t(lim) equation is not a true hyperbola. The effect of the range of t(lim), used in the calculation of the slope of the Wlim-t(lim) relationship (called critical power) are discussed. When critical power is calculated from short supramaximal exercises, this power is higher than the power output which corresponds to a lactate steady state (or an oxygen uptake steady state) and does not correspond to a power output which can be sustained a long time. The authors present experimental data collected during local (knee extension) and general (running and cycling) exercises which suggest that critical power could correspond to a steady state provided that critical power is calculated from heavy submaximal exercises only (t(lim) ranging between 6 and 30 min). It is difficult to predict exhaustion time from critical power or critical velocity because of the hyperbolic nature of the power-t(lim) relationship. On the other hand, a large error in the measure of t(lim) should have a small effect on the calculation of critical power or velocity. In contrast, the value of Y intercept of the Wlim-t(lim) (or Dlim-(t(lim)) relationship should be sensitive to errors in t(lim).

Is the exhaustion time at maximal aerobic speed an index of aerobic endurance?

The velocity of the last stage of the Montreal Track Test (MTT) has been measured in fifteen well trained runners. This velocity (vMTT) was assumed to be close to maximal aerobic running speed. In three different sessions, the subjects ran up to exhaustion at velocities corresponding to 95, 100 and 105% vMTT. The exhaustion time at 100 % vMTT (tlim100) was assumed to be an estimation of the exhaustion time corresponding to maximal aerobic speed. The relationship between exhaustion time (tlim) and distance (Dlim) in the case of running exercises at constant velocity until exhaustion can be described by a linear relationship (Dlim = D + b*tlim). The slope of the relation corresponds to a velocity (vcrit) which can be sustained for a long time. The values of vcrit were calculated from the results of running exercises performed at 95, 100 and 105% of vMTT. The present study showed that tlim at 100% vMTT (tlim100) was negatively correlated with vMTT and vcrit but that D and ratio vcrit/vMTT were independent of vMTT. A theoretical study based on models previously proposed for oxygen kinetics during supramaximal exercises (exponential model and Margaria's model) demonstrates that this negative relationship between vMTT and tlim100 can be explained by the kinetics of the accumulation of oxygen deficit (O2 def). tlim100 should also depend on VO2max, maximal oxygen deficit (Max O2 def) and the relative importance of anaerobic energy when a VO2 plateau is reached. Moreover, the value of tlim100 largely depends on the accuracy of the assessment of vMTT. Consequently, the exhaustion times corresponding to the different estimations of maximal aerobic speed on a track or a treadmill cannot be considered as valid indices of aerobic endurance.

Critical velocity of continuous and intermittent running exercise. An example of the limits of the critical power concept.

The relationship between exhaustion time (tlim) and distance Dlim for running exercises at constant velocity until exhaustion can be described by a linear relationship (Dlim = a + b tlim) whose slope corresponds to a critical velocity. Seven runners participated to the study which compared the critical velocity of continuous versus intermittent running exercises. The critical velocity for continuous running (Vcritc) was calculated from the results (tlimc and Dlimc) of running exercises performed at 95 and 105% of the final velocity of the Montreal Track Test (vMTT). The intermittent running consisted of repetitions of running exercises performed at 95 and 105% vMTT during a time equal to half the value of the corresponding tlimc. The subjects recovered during a time equal to running time while jogging at a slow pace. The critical velocity for intermittent running (Vcriti) was calculated from the cumulated running distance (Dlimi) and cumulated running time (tlimi) corresponding to 95 and 105% vMTT. Vcriti was equal to Vcritc (4.56 +/- 0.444 m.s-1 vs 4.60 +/- 0.416 m.s-1). Nevertheless, in some subjects, the repetition numbers were very different for the intermittent running exercises at 95 and 105% vMTT. This paradoxical result could be explained by the fact that the value of Vcrit should be theoretically little sensitive to a large error in the value of tlim corresponding to a velocity slightly higher than critical velocity, for intermittent exercises as well as continuous exercises.