HumBug – An Acoustic Mosquito Monitoring Tool for use on budget smartphones

Sinka, Marianne; Zilli, Davide; Li, Yunpeng; Kiskin, Ivan; Kirkham, D.; Rafique, Waqas; Wang, Lawrence; Chan, Henry; Gutteridge, Benjamin; Herreros-Moya, Eva; Portwood, Henry; Roberts, Stephen; Willis, Katherine J.

doi:10.1111/2041-210x.13663

Cited by 19 publications

(14 citation statements)

References 48 publications

(71 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…[10] Distributed sensors picking up the wingbeat oscillations from insects are also reported. [11,12] The fact that free-flying insects appear sparsely in time and space, [13] implies that insect monitoring becomes particularly efficient when the probe-volume is elongated over hundreds of meters and trans-illuminated by a laser beam. [14,15] Hereby, an enormous amount of insects can be observed daily using only a couple watts of light as opposed to hyperspectral push broom imaging, [16] where hundreds of Watts illumination is split up between every pixel footprint.…”

Section: Introductionmentioning

confidence: 99%

Remote Nanoscopy with Infrared Elastic Hyperspectral Lidar

Müller

Manefjord

et al. 2023

Advanced Science

View full text Add to dashboard Cite

Monitoring insects of different species to understand the factors affecting their diversity and decline is a major challenge. Laser remote sensing and spectroscopy offer promising novel solutions to this. Coherent scattering from thin wing membranes also known as wing interference patterns (WIPs) have recently been demonstrated to be species specific. The colors of WIPs arise due to unique fringy spectra, which can be retrieved over long distances. To demonstrate this, a new concept of infrared (950-1650 nm) hyperspectral lidar with 64 spectral bands based on a supercontinuum light source using ray-tracing and 3D printing is developed. A lidar with an unprecedented number of spectral channels, high signal-to-noise ratio, and spatio-temporal resolution enabling detection of free-flying insects and their wingbeats. As proof of principle, coherent scatter from a damselfly wing at 87 m distance without averaging (4 ms recording) is retrieved. The fringed signal properties are used to determine an effective wing membrane thickness of 1412 nm with ±4 nm precision matching laboratory recordings of the same wing. Similar signals from free flying insects (2 ms recording) are later recorded. The accuracy and the method's potential are discussed to discriminate species by capturing coherent features from free-flying insects.

show abstract

Section: Introductionmentioning

confidence: 99%

Remote Nanoscopy with Infrared Elastic Hyperspectral Lidar

Müller

Manefjord

et al. 2023

Advanced Science

View full text Add to dashboard Cite

show abstract

“…The study came to an end on 15 August 2021 and the nal set of mosquito sound data uploads were obtained on 16 August 2021. The mosquito sound data recordings that are received and stored in a MongoDB database, were accessed via a dashboard that showed device recording identi cation number as well as visual representation of the audio [47].…”

Section: Milola: Monetary Incentives Onlymentioning

confidence: 99%

“…Over the past ve years, we have developed a novel mosquito survey tool called (HumBug), which uses a budget smartphone running our MozzWear app to record host-seeking mosquitoes by their whining sound and upload the audio data to our servers where an algorithm pipeline detects and identi es host-seeking mosquitoes using the acoustic signature of their ight tone. The HumBug system has the potential to generate unprecedented levels of urgently needed high-quality, spatially accurate mosquito occurrence data without incurring any risk to those conducting the surveys [43][44][45][46][47][48]. By combining these data with environmental variables from remote sensing satellite data, our aim is to develop an innovative system that creates near real-time dynamic heat maps of their distribution, providing information to i) National Malaria Control Programmes (NMCPs) to improve strategies for effective malaria vector control, ii) health policymakers to help develop targeted evidence-based policies for malaria vector management and, iii) the academic community, providing unprecedented levels of occurrence, behavioural and ecological mosquito data in least developed countries (LDCs).…”

Section: Introductionmentioning

confidence: 99%

What incentives encourage local communities to collect and upload mosquito sound data by using smartphones? A case study in Tanzania

Dam

Mponzi

Msaky

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Background: To detect and identify mosquitoes using their characteristic whining sound, we have developed a smartphone application that records the acoustic signature of these sounds, along with the time and location. This data is then sent remotely to a server where algorithms identify the species according to their distinctive acoustic signature. Whilst this system works well, a key question that remains is what mechanisms will lead to effective uptake and use of this mosquito survey tool? We addressed this question by working with local communities in rural Tanzania and providing three alternative incentives: money only, Short Message Service (SMS) reminders and money, and SMS reminders only. We also had a control group with no incentive. Methods: A multi-site, quantitative empirical study was conducted in four villages in Tanzania from April 2021 to August 2021. Consenting participants were recruited and placed into one of the three intervention arms: monetary incentives only; SMS reminders with monetary incentives; and SMS reminders only. There was also a control group (no intervention). To test effectiveness of the mechanisms, the number of audio uploads to the server of the four trial groups on their specific dates were compared. Qualitative focus group discussions and feedback surveys were also conducted to explore participants’ perspectives on their participation in the study and to capture their experiences of using the HumBug sensor.Results: Qualitative data analysis revealed that for many participants (37 out of 81), the main motivation expressed was to learn more about the types of mosquitoes present in their houses. Results from the empirical study supported this finding and revealed that the provision of monetary incentives and/or sending SMS reminders did not appear to encourage greater number of audio uploads when compared to the control. Conclusion: Knowledge on the presence of harmful mosquitoes was the strongest motive for local communities to collect and upload mosquito sound data via the HumBug sensor in rural Tanzania. This finding suggests that most efforts should be made to improve flow of real-time information back to the communities on types and risks associated with mosquitoes present in their houses.

show abstract

“…Its design is based on traditional rectangular bednets found across the malaria-endemic world. The bednet is adapted by the addition of a second outer canopy and a detachable pocket [38]. The pocket is placed at the highest point of the outer canopy above the occupant's head and holds a budget smartphone running the MozzWear app (Figure 2b).…”

Section: Mozzwearmentioning

confidence: 99%

“…As part of the HumBug project, we have developed an end-to-end pipeline to autonomously record, detect and archive mosquito sound. Our pipeline utilises conventional microphones that are found in low-cost mobile phones, and simple adaptions to bednets already commonly used in malaria-endemic areas [38]. This allows broad participation, and the possibility of providing a method for widespread detection in people's homes.…”

Section: Introductionmentioning

confidence: 99%

Automatic Acoustic Mosquito Tagging with Bayesian Neural Networks

Kiskin

Cobb

Sinka

et al. 2021

Machine Learning and Knowledge Discovery in Databases. Applied Data Science Track

Self Cite

View full text Add to dashboard Cite

Deep learning models are now widely used in decision-making applications. These models must be robust to noise and carefully map to the underlying uncertainty in the data. Standard deterministic neural networks are well known to be poor at providing reliable estimates of uncertainty and often lack the robustness that is required for real-world deployment. In this paper, we work with an application that requires accurate uncertainty estimates in addition to good predictive performance. In particular, we consider the task of detecting a mosquito from its acoustic signature. We use Bayesian neural networks (BNNs) to infer predictive distributions over outputs and incorporate this uncertainty as part of an automatic labelling process. We demonstrate the utility of BNNs by performing the first fully automated data collection procedure to identify acoustic mosquito data on over 1,500 hours of unlabelled field data collected with low-cost smartphones in Tanzania. We use uncertainty metrics such as predictive entropy and mutual information to help with the labelling process. We show how to bridge the gap between theory and practice by describing our pipeline from data preprocessing to model output visualisation. Additionally, we supply all of our data and code. The successful autonomous detection of mosquitoes allows us to perform analysis which is critical to the project goals of tackling mosquito-borne diseases such as malaria and dengue fever.

show abstract

HumBug – An Acoustic Mosquito Monitoring Tool for use on budget smartphones

Cited by 19 publications

References 48 publications

Remote Nanoscopy with Infrared Elastic Hyperspectral Lidar

Remote Nanoscopy with Infrared Elastic Hyperspectral Lidar

What incentives encourage local communities to collect and upload mosquito sound data by using smartphones? A case study in Tanzania

Automatic Acoustic Mosquito Tagging with Bayesian Neural Networks

Contact Info

Product

Resources

About