Biomedical subjects

J Vermont

Publications and source records attributed to J Vermont.

5 recordsLinked to original sources

PubMed1991

Formulas for threshold computations.

Given a continuous variable S, which density functions on two subgroups omega + and omega - of a population omega are known (with for instance a higher mean value on omega + than on omega -), we first define two strategies for classification in these groups; the first one (MWC) consists in determining a threshold alpha such that classifying in omega + when S greater than or equal to alpha, in omega - otherwise, leads to the highest percentage of well-classed elements. The second one consists in choosing the most probable group, given the observed value of S. We give mathematical formulas for the thresholds involved in these two strategies when the density functions, determined by the application of the maximum entropy principle, are those of normal distributions. These formulas prove that the two considered strategies are frequently equivalent, and we give simpler formulas when the partial variances of S on omega + and omega - are unknown or approximately equal. All the formulas are adapted to the case where a cost coefficient is introduced to display the unequal seriousness of the two possible errors (misclassification in omega + or omega -). Then we consider an example, where we see that the computed thresholds can be graphically validated from empirical curves and have the same performances on the learning sample and on a test sample.

Analysis of Variance↗

PubMed1991

Strategies for graphical threshold determination.

Determining a threshold for a quantitative variable (arising in biological measurements for instance) is a common problem in medical decision making. We define seven commonly used strategies: each one leads to an optimal determination. To these strategies correspond relevant empirical curves: the ROC curve for strategies involving the sensitivity or the specificity, the predictive ROC curve (P-ROC curve) for strategies involving the positive and negative predicting values, and the well classified frequencies curve (WCF curve) for classification strategies where all misclassifications have the same importance. For one of the considered strategies, there also exists a theoretical formula for the optimal threshold, elicited within a classical probabilistic model, which gives a considerable advantage to this strategy. These strategies are applied to a stimulated example containing 702 cases, where we see that they lead to different optimal threshold values. Finally, we briefly review a practical application in the determination of thresholds for glycemia measurements, leading to the choice of one of them as the optimal one to consider in the gestational diabetes mellitus prediction.

Diagnosis, Computer-Assisted↗

PubMed1991

Automatic classification of cells in cell cycle phases based on Ki-67 antigen quantification by fluorescence microscopy.

Ki-67 antigen is thought to be a marker of cell proliferation, as it can be detected in cycling cells, i.e. cells in G1, S, G2 and M phases, but not in resting cells. The immunocytochemical staining pattern obtained by the Ki-67 monoclonal antibody varies, depending on the cell cycle phases. Analysis of double staining of Ki-67 antigen and DNA in the MCF-7 cell line by videomicrofluorometry allows the description of both the level and the pattern of Ki-67 staining in the form of a set of parameters defining each cell. These parameters were measured in MCF-7 cell populations characterized according to their position in the cell cycle. They were submitted to a statistical analysis (principal component and discriminant analysis) which allowed the determination of the optimal parameters to characterize a given cellular group and permitted the use of these parameters for an automatic classification of cells in the different cell cycle phases. In G1, S, G2, prophase + metaphase and anaphase + telophase cells, these parameters allowed a classification of cells with a good-classification rate of 94.37%. A comparison of this method with methods based on the DNA histogram and bromodeoxyuridine uptake was performed. The classification coefficients stemming from the discriminant analysis were introduced into a program to obtain, automatically, the Ki-67 labelling index and the percentages of cells in each phase. This method, which allows a quick evaluation of the proliferation and the phase indices, may be more widely applicable.

Breast Neoplasms↗

PubMed1985

[Role of acute alcoholism in violent death. Statistics of the Lyon Medico-Legal Institute (1981-2)].

Over a 2-year period (1981-1982), blood alcohol levels were measured in 1371 post-mortem examinations performed at the medico-legal Institute of Lyon. These levels were higher in men, in subjects who died between 9 p.m. and 3 a.m., in corpses found in water and in subjects killed by a fall or in a fire. Among the 4 main types of death (homicide, suicide, accident and natural death), those by homicide had the highest, and those by suicide the lowest blood alcohol concentrations.

Accidents↗

PubMedDate not supplied

Ultrastructural changes associated with peripheral neuropathy in HIV/AIDS.

Light and electron microscopic studies were performed on neuromuscular biopsy specimens from 12 neurologically affected seropositive patients, 7 with the acquired immune deficiency syndrome (AIDS), 2 with AIDS-related complex, and 3 with no symptoms except for neuropathy. All patients had an axonal injury associated or unassociated with demyelination and peripheral neurogenic atrophy. Capillary lesions were consistently present, which seems to be a new finding. Moreover, tubuloreticular inclusions (TRIs) were found in endomysial (9 of 12 cases) and endoneurial (7 of 12 cases) vessels. Statistical analysis showed that TRIs in more than 20% of endomysial vessels correlated with a survival time shorter than 12 months (P = 0.028). Thus the quantification of TRIs seems to be one of the vital prognostic elements in this patient population.

Acquired Immunodeficiency Syndrome↗