Introduction
In 1955, at the Dartmouth Summer Research Project on Artificial Intelligence, John McCarthy coined AI. Artificial Intelligence (AI) refers to a machine’s ability to learn and demonstrate Intelligence superior to humans and animals. Our personal and social lives have been changed with the influx of Artificial Intelligence. Speech recognition, facial recognition, game AI, intelligent voice assistants, and self-driving vehicles are all examples of AI in use today. In medicine, AI has proven to be effective in areas such as radiology, cardiology, and oncology. AI aids in extracting information from medical images, which is critical in clinical diagnosis and therapy. Machine learning (ML), natural language processing (NLP), and robotic surgery are the three types of AI used in medical applications.
Vector-borne diseases are responsible for 17% of the estimated global burden of infectious diseases. It causes more than 700,000 deaths yearly, and at least 80% of the global population lives in areas at risk. Artificial Intelligence used in the Entomological characterization of insects is a progressive action towards controlling and managing vector-borne disease. The World Health Organization prioritizes entomology research to develop methods that can be used to minimize the incidence and mortality of vector-borne diseases and avert epidemics around the world. The process in the entomological characterization of an adult mosquito begins with the assembly of the mosquito on an entomological plan followed by observing the specimen under the microscope, evaluating the morphological characters, and finally, the species is classified. Public health workers in many parts of the world still use the traditional visual examination method to identify mosquito species and sex. However, the above practice is time-consuming, tiring, and requires several years of experience. Also, during the process, some of the mosquito’s samples are damaged and lose morphological characteristics. Transporting the mosquito sample from field to lab also becomes a challenge as it can dry the mosquito’s body or change its color. Thus, AI helps entomologists, epidemiologists, and public health professionals capture and identify the insect species and its sex quickly and help map and organize control measurements where the transmission rate is high. With the help of Machine learning (ML) and deep learning techniques, AI helps classify mosquitoes in their different phases of the life cycle.
Benefits of AI application in the control and management of vector-borne disease
Artificial Intelligence (AI) increases the participation of communities to control and reduce the burden of recurrent epidemics caused by vector-borne disease. AI is used to
- Enhance the performance of Mosquito Traps
Artificial Intelligence helps to design mosquito traps by incorporating advanced functions. For example, it enables the mosquito traps to target a specific category of mosquito and leave the other types. Furthermore, scientists may use artificial Intelligence (AI) to collect and store vital data to understand mosquito behavior better and correlate data such as the date and time of capture, the species caught, and environmental data (humidity and temperature). Machine learning has brought a revolution in designing advanced mosquito traps that uses laser sensor and audio analysis techniques to recognize the particular category of mosquito and have an accuracy of 98%. - Classify the type of mosquito responsible for causing the disease
Identifying the correct mosquito is crucial in developing a control and management plan to curb vector-borne diseases. We have discussed above the challenges faced in the traditional practice of collecting and identifying the type of mosquito. A method that was developed to solve the problems faced in conventional practice was the introduction of molecular techniques like DNA Barcodes. However, those molecular techniques are slow and expensive.
Artificial Intelligence (AI) has become the answer to the above challenges. AI helps automate the mosquito classification process, prevents damage to the morphological characteristics of the mosquito’s body, and reduces the need for trained and experienced personnel to identify the type and family of the mosquito visually.
AI has been found to have an accuracy of 86%-100%, inefficiently classifying mosquitoes responsible for vector-borne disease. Support vector machine (SVM) techniques and digital image processing use AI technology to detect mosquitoes like Aedes aegypti, Aedes albopictus, and other vectors responsible for diseases like dengue, malaria, chikungunya, etc. Support vector machine (SVM) techniques and digital image processing take a photo of the insect and then compares it with the data. The instrument is designed and programmed to respond only if the particular mosquito species, e.g., Aedes aegypti, is found in the captured sample. The accuracy of such AI-based techniques is nearly 90%.
Digital Images of Mosquito larva have been used to identify the Aedes species of the mosquito larvae with the help of a Machine learning Algorithm. The results are 100%accurate for identifying Aedes larva; however, the misclassification rate was 30% for other mosquitoes.
Conclusion
Artificial Intelligence has been promising towards the growth and development of medical sciences. However, it is an essential field that needs to be merged with medical science to control and manage different life-threatening diseases. Artificial Intelligence is also used in Thermal cameras to capture temperature, artificial immune recognition systems to diagnose various diseases, epidemiological prediction models to carry out clinical and population epidemiology studies, reproductive disorders to identify the cause of infertility, and improve Assisted Reproductive Technology (ART) techniques to cure infertility. With the advancement and merging of Artificial Intelligence in modern medicine, who knows, in the future, we may identify the species of the mosquito and the diseases associated with them just by clicking a photo of the mosquito through our smartphones.
References
- Deo RC. Machine learning in medicine. Circulation. 2015 Nov 17;132(20):1920-30.
- Wang, R., Pan, W., Jin, L., Li, Y., Geng, Y., Gao, C., Chen, G., Wang, H., Ma, D. and Liao, S., 2019. Artificial Intelligence in reproductive medicine. Reproduction, 158(4), pp.R139-R154.
- Pal NR, Pal SK. A review on image segmentation techniques. Pattern recognition. 1993 Sep 1;26(9):1277-94.
- Darcy AM, Louie AK, Roberts LW. Machine learning and the profession of medicine. Jama. 2016 Feb 9;315(6):551-2.
- Agrebi S, Larbi A. Use of artificial Intelligence in infectious diseases. InArtificial Intelligence in precision health 2020 Jan 1 (pp. 415-438). Academic Press.
- da Silva Motta D, Badaró R, Santos A, Kirchner F. Use of Artificial Intelligence on the Control of Vector-Borne Diseases. Vectors and Vector-Borne Zoonotic Diseases. 2018 Nov 5.