Decision trees based on automatic learning and their use in cardiology

Computerized information systems, especially decision support systems, have become an increasingly important role in medical applications, particularly in those where important decision must be made effectively and reliably. But the possibility of using computers in medical decision making is limited by many difficulties, including the complexity of conventional computer languages, methodologies and tools. Thus a conceptual simple decision making model with the possibility of automating learning should be used. In this paper we introduce a cardiological knowledge-based system based on the decision tree approach supporting the mitral valve prolapse determination. Prolapse is defined as the displacement of a bodily part from its normal position. The term mitral valve prolaps (PMV), therefore, implies that the mitral leaflets are displaced relative to some structure, generally taken to be the mitral annulus. The implications of the PMV are the following: disturbed normal laminar blood flow, turbulence of the blood flow, injury of the chordae tendinae, the possibility of thrombus's composition, bacterial endocarditis, and finally hemodynamic changes defined as mitral insufficiency and mitral regurgitation. Uncertainty persists about how it should be diagnosed and about its clinical importance. It is our deep belief that the echocardiography enables properly trained experts armed with proper criteria to evaluate PMV almost 100%. But unfortunately, there are some problems concerned with the use of echocardiography. In that manner we have decided to start a research project aimed at finding new criteria and enabling the general practitioner to evaluate PMV using conventional methods and to select potential patients from the general population. To empower one to perform needed activities we have developed a computer tool called ROSE (computeRised prOlaps Syndrom dEtermination) based on algorithms of automatic learning. This tool supports the definition of new criteria and the selection of potential PMV-patients.     

INVOLVEMENT OF NON-NMDA AND NMDA RECEPTORS IN GLUTAMATE-INDUCED PRESSOR OR DEPRESSOR RESPONSES OF THE PONS AND MEDULLA

1. Fifty-five intact and six baroreceptor denervated and vagotomized cats of either sex were anaesthetized intraperito-neally with urethane (400 mg/kg) and a-chloralose (40 mg/kg). Responses of the systemic arterial pressure (SAP), mean SAP (MSAP) and sympathetic vertebral nerve (VNA) and renal nerve activities (RNA) were recorded. 2. In intact animals, monosodium L-glutamate (Glu, 0.1 mol/L, 50 nL) was microinjected into pressor areas of the locus coeruleus (LC), gigantocellular tegmental field (GTF), rostral ventrolateral medulla (RVLM) and dorsomedial medulla (DM), and the depressor areas of caudal ventrolateral medulla (CVLM). The induced actions were compared before and after microinjection of either glutamate antagonists, glutamate diethylester (GDEE, 0.5 mol/L, 50–100nL), a competitive AMPA receptor blocker, or 2-amino-5-phosphonovaleric acid (D-AP5, 0.025 mol/L, 50–100 nL), a competitive N-methyl-D-aspartate (NMDA) receptor blocker. GDEE completely blocked the increases of SAP and VNA elicited from all pressor areas. D-AP5 only partially blocked the pressor but slightly blocked VNA and RNA responses from LC, GTF and DM, particularly those from RVLM. Neither GDEE nor D-AP5 blocked the depressor responses of SAP and two nerve activities elicited from CVLM. 3. In baroreceptor denervated animals, NMDA (2 mmol/L, 50–100 nL) and AMPA (0.2 mmol/L, 50–100 nL) were micro-injected into the same pressor areas of GTF, RVLM and DM and the depressor area of CVLM responsive to Glu activation (0.1 mol/L, 30 nL). In RVLM, DM and CVLM, the results of either NMDA or AMPA were similar to those induced by Glu. However, in GTF, microinjection of either NMDA or AMPA did not induce similar responses to Glu. This suggests that the nature of GTF may differ from RVLM and DM. 4. The above results suggest that the Glu-induced pressor responses from LC, GTF, DM and especially RVLM, are primarily mediated through AMPA receptors. The Glu-induced depressor responses from CVLM may not be predominantly mediated by either AMPA or NMDA receptors. 5. In both baroreceptor-intact and -denervated cats stimulation of the pressor areas often produced an increase of VNA and a decrease of RNA, while in the depressor CVLM decreased both VNA and RNA. The VNA, but not RNA were positively correlated with the pressor responses, while both VNA and RNA were positively correlated with the depressor responses. This may suggest that neurons of the sympathetic vertebral and renal nerves are topographically organized in the brain.     

Proximal femoral focal deficiency: radiologic analysis of 49 cases

Proximal femoral focal deficiency, an uncommon congenital anomaly, necessitates early radiologic classification for surgical planning and treatment. Objective radiographic criteria, including femoral length index, acetabular depth index, acetabular angle index, and shape of the proximal femur were determined in 49 patients before cartilaginous ossification of the femoral capital epiphysis; final classification was based on follow-up radiographs or findings at arthrography or surgery. These parameters were analyzed to determine the accuracy and contributions of each in classification. Correct classification into one of three groups was possible in 86% of cases with use of three of the parameters: femoral length index, acetabular depth index, and shape of the proximal femur. The acetabular angle was found to contribute insignificantly to classification. Magnetic resonance imaging, used in only one case, depicted the nonossified cartilaginous femoral capital epiphysis, thus obviating the need for invasive diagnostic procedures and facilitating early classification.