There is still no viable way for prescreening of ovarian cancer. This could be accomplished using a micropipette thermal sensor (MPTS) to find the thermal properties of a cell [1-3]. These thermal properties can be used as bio markers to identify cancerous cells [4]. However, the MPTS is measuring the change in temperature of a cell; the properties such as specific heat, density, and thermal conductivity can be calculated using a partial differential equation (PDE) of the heat diffusion equation. In practical conditions such as surgery the calculations could not be done realistically in that time constraint. Therefore, it was proposed for an Artificial Neural Network (ANN) to be trained to solve for these properties [5]. Machine learning make it possible for machines to process large amounts of data to accomplish specific tasks and recognize patterns in the data. This is useful in applications such as computer vision, speech processing, and game playing [6]. ANNs use a system of neurons with specified weights and a bias, this system is known as a perception [7]. As the system is trained these weights are altered to minimize an error equation, this is known as backpropagation [8]. The research will cover training of the ANN to solve for three parameters in the heat diffusion equation; density, thermal conductivity, and specific heat. An Arduino programmed with the trained ANN will process the data and return values for the proposed properties. Expectations are these properties will vary only slightly and the error between the known and returned would be comparable to minimum error during backpropagation.