Digital nose for healthcare: Diagnosing diabetes and heart diseases via a low-cost digital nose

Title of the project

Digital nose for healthcare: Diagnosing diabetes and heart diseases via a low-cost digital nose.

Problem statement

Non-invasive diagnosis of different lifestyle diseases like diabetes and heart disease for the common population by breath analysis. The detection of the different biomarkers in human breath related to different diseases. This project will also enable early low-cost detection of Diabetic ketoacidosis, which is a serious metabolic condition that occurs when the body produces high levels of ketones and the pH of the blood becomes too acidic. Ketoacidosis is most commonly seen in people with uncontrolled diabetes, but it can also occur in alcoholics and other individuals with certain medical conditions. If left undetected, this condition is life-threatening and warrants urgent hospital care.

Project brief

Digital noise (d-nose) is an IoT device that detects smell. D-nose may have a number of health-related applications, especially with respect to low-cost and early detection of a number of diseases like diabetes, and heart and lung diseases. Here, a number of researchers have shown that certain biomarkers in the human body are related to the incidence of certain diseases. For example, in diabetic patients, body cells cannot metabolize the glucose in the blood properly. In such patients, when the liver breaks down fat for energy, ketones like acetone are produced. As acetone it is volatile, it can be detected in the exhaled breath. Prior research has shown that the concentration of acetone in breath is correlated with BGLs. As a result, the analysis of components in breath can be used as a non-invasive approach for blood-glucose level (BGL) monitoring diabetic patients. Based upon this example and other biomarkers of diseases, in this project, the PI and Co-PI plan to develop a low-cost digital nose for the non-invasive BGL monitoring of BGL patients. Thus, the device is supposed to not only predict a diabetic from a non-diabetic, but it is also going to assess the BGL level in mmlo/l, where levels greater than 7.11 is an indication of hyperglycemia or diabetes. Furthermore, depending on the availability of patients, we will also try to predict the incidence of cardiopathy (i.e., heart disease) among patients. The cardiopathy investigation would be first done with respect to the presence or absence of heart disease and later, based upon the availability of patients, may be extended to the type of heart disease. In this case, the data can come from patients who have been recently treated, or who are at high-risk and will be labeled based on the other standard clinical parameters.

Technologies involved?

  • Sensors: Digital noses use various types of sensors to detect and measure the chemical compounds present in a sample. The most common types of sensors used in e-noses include metal oxide sensors, conducting polymer sensors, quartz crystal microbalance sensors, and optical sensors.
  • Signal processing: Once the sensors detect the presence of different volatile compounds, the signals from the sensors are processed and analysed to identify the specific odors. This involves using algorithms and statistical models to identify unique patterns and signatures for different odors.
  • Machine learning: Machine learning techniques are often used in conjunction with signal processing to improve the accuracy of odor detection and identification. This involves training the digital nose on a large dataset of known odors and their corresponding sensor signals, so it can learn to recognize and identify new odors.
  • Data visualization: The results of digital nose measurements can be displayed visually in various formats to aid in interpretation. This may involve generating 3D plots, heatmaps, or other graphical representations of the sensor data.
  • Software and hardware integration: To develop a functional digital nose, it is necessary to integrate the various hardware and software components into a cohesive system. This involves designing and building the sensor array, developing the signal processing and machine learning algorithms, and integrating the various components into a user-friendly interface.

Solution is relevant to which industry

  • Food and beverage industry: Digital noses can be used to detect and identify the aroma compounds in foods and beverages, which can help food manufacturers develop new products, ensure product consistency, and monitor the quality of their products.
  • Environmental monitoring: Digital noses can be used to monitor air quality and detect the presence of harmful or noxious gases in the environment. This can be useful in a variety of settings, including industrial facilities, waste management facilities, and agricultural operations.
  • Healthcare industry: Digital noses can be used in medical diagnostics to detect and identify the volatile organic compounds (VOCs) present in breath samples, which can provide early detection of diseases such as lung cancer, diabetes, and other metabolic disorders.
  • Cosmetics industry: Digital noses can be used to evaluate the fragrance of cosmetic products, such as perfumes and deodorants, and ensure that they meet the desired specifications.
  • Military and security: E-noses can be used in military and security applications to detect explosives, chemical weapons, and other hazardous materials.

Overall, digital noses have the potential to be used in a wide range of industries where odor detection and identification are important, and their applications are continually expanding as technology advances.