Page 179 - Kaleidoscope Academic Conference Proceedings 2021

IDENTIFICATION OF DEADLIEST MOSQUITOES USING WING BEATS SOUND
CLASSIFICATION ON TINY EMBEDDED SYSTEM USING MACHINE LEARNING AND
EDGE IMPULSE PLATFORM

Dr. Kirankumar Trivedi; Harsh Shroff

Vishwakarma Government Engineering College, Ahmedabad, India

ABSTRACT Culex, as seen in Figure 1.
Mosquitoes are the deadliest animal on the planet, infecting
about 700 million people each year and causing over one
million deaths, accounting for 17% of all infectious illnesses
worldwide. We are still fighting the three most deadly
mosquito species, Anopheles, Aedes, and Culex, 124 years
after Sir Ronald Ross made the first pivotal discovery.
Mosquitoes are difficult to detect manually since they are
small and fly rapidly. The auditory categorization of Figure 1 – Anopheles, Aedes, and Culex mosquitoes
mosquito wing beats may be used to detect them using Source: http://www.bigstockphoto.com
machine learning. This article discusses an Arduino Nano
BLE 33 Sense–based prototype that collects audio data from Table 1 – Vector–borne diseases, according to the mentioned
mosquito wing beats and utilizes TinyML to automatically mosquito species[2]
classify mosquito species. With 88.3% accuracy, the TinyML
Vector Disease caused Type of pathogen
system developed by Edge Impulse based on the HumBug
Aedes Chikungunya Virus
project mosquito wing beats dataset recognizes mosquito
Dengue Virus
types. To conduct this research, the frequency of mosquito
Lymphatic filariasis Parasite
wing beats was graphically represented as a feature using a
Rift Valley fever Virus
spectrogram. Furthermore, live mosquito detection studies
Yellow Fever Virus
using the low–cost Arduino Nano BLE 33 Sense yielded
Zika Virus
excellent results. During testing, the model had an accuracy
Anopheles Lymphatic filariasis Parasite
of 88.3% and a loss of 0.26. The use of machine learning
Malaria Parasite
to solve the challenge of manual mosquito type identification
Culex Japanese encephalitis Virus
is efficient and has the potential to have a large impact on
Lymphatic filariasis Parasite
vector–borne illness management. The model may still be
West Nile fever Virus
fine–tuned to get more accurate results with reduced latency.
In addition, the deployment went as expected.
Aedes aegypti mosquitos, according to a WHO study, are
Keywords - Acoustic Classification, Arduino Nano BLE 33
excellent vectors of potentially serious illnesses such as
Sense, Edge Impulse, machine learning, mosquito
dengue fever, chikungunya fever, yellow fever, and Zika
virus[4]. It has been hypothesized that Aedes aegypti is a
1. INTRODUCTION
possible vector of Venezuelan Equine Encephalitis virus[5],
and vector competence tests have revealed that Aedes aegypti
The mosquito is possibly the only predator in human history
may transmit West Nile virus.
to have survived for millennia, inflicting death and destruction
through a variety of vector–borne illnesses, particularly
malaria[1]. In reality, the mosquito was discovered to be the Another WHO report says that malaria is a life–threatening
world’s deadliest animal, killing over 700,000 people each parasitic illness spread by bites from infected female
year[3], according to a list of the world’s deadliest creatures. Anopheles mosquitoes. Malaria affected an estimated 229
Sir Ronald Ross discovered that Anopheles mosquitoes were million people globally in 2019. It was estimated that
responsible for the spread of the malaria parasite on August 409,000 people died from malaria, the same year[6]. The
20, 1897. Every year on 20 August, World Mosquito Day malaria–carrying Anopheles mosquito might be a source of
is celebrated to honor this pivotal finding in the fight against additional brain tumor viruses. The connection between
malaria. malaria outbreaks in the United States and reports of brain
Table 1 shows a list of vector–borne illnesses caused by the tumor incidence by state provides evidence of a correlation
mosquito species listed above, namely Anopheles, Aedes, and between anopheles and brain tumors[7].

978-92-61-33881-7/CFP2168P @ ITU 2021 – 117 – Kaleidoscope