Analysis of thermal properties of wheelchair cushions with thermography M. Ferrarin 1 N. Ludwig 2 1 Centro di Bioingegneria, Fondazione Don Carlo Gnocchi IRCCS-Politecnico di Milano, Milan, Italy 2 Istituto di Fisica Generale Applicata, Universita Á degli Studi, Milan, Italy AbstractÐThermal properties of wheelchair cushions have been traditionally studied with thermistor probes, which provide temperature values of limited areas (spot analysis). In this paper, we describe a novel procedure based on thermography for assessing the distribution of temperature over the entire surface of wheelchair cushions. The thermal transient during contact with the body (heating phase) and after use (cooling phase) is considered. The procedure was tested in four different seat cushions (with a gel pad, air-®lled cells, gel-®lled bubbles and foam-®lled bubbles) used by a normal subject. Observed results were compatible with the predicted outcomes based on an analysis of the materials and structures. Speci®cally: (i) air-®lled cushions exhibited the fastest thermal transi- ents, gel cushions the slowest transients, while cushions with a mixed structure exhibited intermediate behaviour; (ii) cushions made from ¯at surfaces of foam exhibited the highest peak temperatures (30 :8  C) as compared to those with air- ®lled cells (30 :35  C ) or bubble-shaped surfaces (29 :7  C ); (iii) the average tempera- ture under the thighs was signi®cantly higher than that under the ischiatic area in all cushions (29 :6  C compared with 28 :7  C, p < 0 :05). It is shown that the present method can be used to differentiate between different cushions. Although the `macro-analysis' inherent in thermography appears to be suited for improving cushion design, this approach should be further investigated to determine its reliability. KeywordsÐWheelchair cushions, Thermography Med. Biol. Eng. Comput., 2000, 38, 31±34 1 Introduction IT HAS been suggested (FISHER et al., 1978) that factors such as moisture and high temperature may increase susceptibility to pressure sores in wheelchair±bounded individuals. A rise in skin temperature may increase the metabolic demands of the cells in a region already suffering from poor circulation due to pressure. Moreover, at elevated skin temperatures perspiration increases. A study by STEWART et al. (1980) demonstrated that the tensile strength of the skin decreased with an increase in its relative humidity, thus possibly increasing mechanical damage from shear stress and the risk of infection. Therefore, methods of determining the thermal properties of wheelchair cushions may be useful in designing better anti±pressure ulcer devices. Relatively few studies have addressed the thermal aspects of wheelchair cushions. FISHER et al. (1978), STEWART et al. (1980) and FINESTONE et al. (1991) used thermistor probes to evaluate the temperature at speci®c points of the cushion-body interface (ischial tuberosities and=or sacral area), with the inherent disadvantage of recording data only from the limited surface under the sensor (typically 51 cm 2 ). The dif®culty in sensor positioning, the large number of sensors needed for a global analysis and the alteration of thermal exchanges between the body and the cushion are the main drawbacks of this traditional approach. Thermographical investigation yields more comprehensive information about the overall surface; therefore, in our opinion, it has the potential to improve the process of cushion design. While thermographic images of speci®c body areas have been used as indicators of tissue damage or the skin vascular response to pressure in a variety of studies (TRANDEL, 1975; PYE et al., 1976; NEWMAN et al., 1981), we know of no studies that have used this technique for the analysis of the cushion itself. The aim of the present study was to develop and test a procedure for detecting the thermal properties of wheelchair cushions using an infrared thermographic system. The proce- dure was designed to enable us to investigate the effect of different fabrication materials on cushion thermal distribution and on the thermal transient both during contact with the body (heating phase) and after raising (cooling phase). The proce- dure was tested on four commercially available pressure- reducing cushions using one normal subject to determine whether the observed outcomes were compatible with the prediction based on an analysis of the materials and structures of the cushions. Correspondence should be addressed to Dr M. Ferrarin; email: ferramau@mail.cbi.polimi.it First received 20 July 1999 and in ®nal form 13 September 1999 ß IFMBE: 2000 Medical & Biological Engineering & Computing 2000, Vol. 38 31