Semi-Supervised Clustering-Based DANA Algorithm for Data Gathering and Disease Detection in Healthcare Wireless Sensor Networks (WSN) DOI Creative Commons
Anurag Sinha, Turki Aljrees, Saroj Kumar Pandey

et al.

Sensors, Journal Year: 2023, Volume and Issue: 24(1), P. 18 - 18

Published: Dec. 19, 2023

Wireless sensor networks (WSNs) have emerged as a promising technology in healthcare, enabling continuous patient monitoring and early disease detection. This study introduces an innovative approach to WSN data collection tailored for detection through signal processing healthcare scenarios. The proposed strategy leverages the DANA (data aggregation using neighborhood analysis) algorithm semi-supervised clustering-based model enhance precision effectiveness of WSNs. optimizes energy consumption prolongs node lifetimes by dynamically adjusting communication routes based on network’s real-time conditions. Additionally, clustering utilizes both labeled unlabeled create more robust adaptable technique. Through extensive simulations practical deployments, our experimental assessments demonstrate remarkable efficacy method model. We conducted comparative analysis efficiency, utilization, accuracy against conventional techniques, revealing significant improvements quality, rapid diagnosis. combined offers WSNs compelling solution responsiveness reliability diagnosis processing. research contributes advancement systems offering avenue improved care, ultimately transforming landscape enhanced capabilities.

Language: Английский

Leveraging the collaborative power of AI and citizen science for sustainable development DOI Creative Commons
Dilek Fraisl, Linda See,

Steffen Fritz

et al.

Nature Sustainability, Journal Year: 2024, Volume and Issue: unknown

Published: Dec. 16, 2024

Language: Английский

Citations

4
Citizen science and negotiating values in the ethical design of AI-based technologies targeting vulnerable individuals DOI Creative Commons
Alessandra Cenci

AI and Ethics, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 30, 2025

Abstract Citizen science is the new mantra both in academic circles and public discourse. While citizen ideal conceptually broad, If how it can be realized fields often depicted as value free/value neutral—such applied AI—is controversial. The practical challenges generating ethical AI encapsulating are addressed by targeting scientific practices underlying participatory design of an AI-based tracking app aimed at enhancing safety wellbeing vulnerable citizens with dementia a Danish municipality through engagement local community. focus on process social construction its rationale: values have been debated, traded-off, selected via participatory-deliberative methods engaging experts non-expert stakeholders scientists. An emphasis import dialogic interaction for negotiating open conversations within diverse groups interest. Deliberative procedures beneficial to produce embodying vital desiderata since users’/citizens' values, needs, expectations fulfilled while technical-efficiency standards also met. result methodology designing that better expresses true spirit liberal democracies (value-laden, pluralistic, inter-disciplinary, inclusive, participatory, cooperative, solidarity-oriented). Hence, trust acceptance generated, even contentious “surveillance” technologies, enhanced digital innovation perceived truly citizens-/humans-centred society-oriented.

Language: Английский

Citations

0
Making sense of fossils and artefacts: a review of best practices for the design of a successful workflow for machine learning-assisted citizen science projects DOI Creative Commons

Isaak Eijkelboom,

Anne S. Schulp, Luc Amkreutz

et al.

PeerJ, Journal Year: 2025, Volume and Issue: 13, P. e18927 - e18927

Published: Feb. 13, 2025

Historically, the extensive involvement of citizen scientists in palaeontology and archaeology has resulted many discoveries insights. More recently, machine learning emerged as a broadly applicable tool for analysing large datasets fossils artefacts. In digital age, science (CS) (ML) prove to be mutually beneficial, combined CS-ML approach is increasingly successful areas such biodiversity research. Ever-dropping computational costs smartphone revolution have put ML tools hands with potential generate high-quality data, create new insights from elevate public engagement. However, without an integrated approach, projects may not realise full scientific engagement potential. Furthermore, object-based data gathering artefacts comes different requirements approaches than observation-based monitoring. this review we investigate best practices common pitfalls interdisciplinary field order formulate workflow guide future palaeontological archaeological projects. Our subdivided four project phases: (I) preparation, (II) execution, (III) implementation (IV) reiteration. To reach objectives manage challenges subject domains (CS tasks, development, research, stakeholder app/infrastructure development), tasks are formulated allocated roles project. We also provide outline online CS platform which will help project’s Finally, illustrate our practice showcase differences more commonly available approaches, discuss LegaSea sand nourishments western Netherlands studied.

Language: Английский

Citations

0
A data science approach to mitigating data challenges in serious gaming DOI Creative Commons
Germain Abdul-Rahman, Noman Haleem, Andrej Zwitter

et al.

Discover Data, Journal Year: 2025, Volume and Issue: 3(1)

Published: March 3, 2025

Language: Английский

Citations

0
Application of geographic information system and remote sensing technology in ecosystem services and biodiversity conservation DOI
Maqsood Ahmed Khaskheli, Mir Muhammad Nizamani,

Umed Ali Laghari

et al.

Elsevier eBooks, Journal Year: 2025, Volume and Issue: unknown, P. 97 - 122

Published: Jan. 1, 2025

Language: Английский

Citations

0
Forensic insights into coastal microplastic pollution: Pathways, source attribution techniques, innovations, and regulatory implications DOI
Azubuike V. Chukwuka,

Dami Emmanuel Omogbemi,

Ayotunde Daniel Adegboyegun

et al.

Regional Studies in Marine Science, Journal Year: 2025, Volume and Issue: unknown, P. 104166 - 104166

Published: April 1, 2025

Language: Английский

Citations

0
Quantifying online citizen science: Dynamics and demographics of public participation in science DOI Creative Commons
Bruno J. Strasser, Élise Tancoigne, Jérôme Baudry

et al.

PLoS ONE, Journal Year: 2023, Volume and Issue: 18(11), P. e0293289 - e0293289

Published: Nov. 21, 2023

Citizen scientists around the world are collecting data with their smartphones, performing scientific calculations on home computers, and analyzing images online platforms. These citizen science projects frequently lauded for potential to revolutionize scope scale of collection analysis, improve literacy, democratize science. Yet, despite attention has attracted, it remains unclear how widespread public participation is, changed over time, is geographically distributed. Importantly, demographic profile participants uncertain, thus what extent contributions helping Here, we present largest quantitative study in based accounts more than 14 million two decades. We find that trend broad rapid growth observed early 2000s since diverged by mode participation, consistent nature sensing, but a decline seen crowdsourcing distributed computing. Most projects, except heavily dominated men, vast majority participants, male female, have background The analysis here provides, first robust 'baseline' describe global trends participation. results highlight current challenges future Beyond presenting our collated data, work identifies multiple metrics examination and, generally, crowds. It also points limits studies capturing personal, societal, historical significance

Language: Английский

Citations

10
Recognizability bias in citizen science photographs DOI Creative Commons
Wouter Koch, Laurens Hogeweg, Erlend B. Nilsen

et al.

Royal Society Open Science, Journal Year: 2023, Volume and Issue: 10(2)

Published: Feb. 1, 2023

Citizen science and automated collection methods increasingly depend on image recognition to provide the amounts of observational data research management needs. Recognition models, meanwhile, also require large from these sources, creating a feedback loop between tools. Species that are harder recognize, both for humans machine learning algorithms, likely be under-reported, thus less prevalent in training data. As result, may hamper mostly species already pose greatest challenge. In this study, we trained models various taxa, found evidence 'recognizability bias', where more readily identified by alike available This pattern is present across multiple does not appear relate differences picture quality, biological traits or metrics other than recognizability. has implications expected performance future with data, including such challenging species.

Language: Английский

Citations

8
Being confident in confidence scores: calibration in deep learning models for camera trap image sequences DOI Creative Commons
Gaspard Dussert, Simon Chamaillé‐Jammes, Stéphane Dray

et al.

Remote Sensing in Ecology and Conservation, Journal Year: 2024, Volume and Issue: unknown

Published: June 16, 2024

Abstract In ecological studies, machine learning models are increasingly being used for the automatic processing of camera trap images. Although this automation facilitates and accelerates identification step, results these may lack interpretability their immediate applicability to downstream tasks (e.g. occupancy estimation) remains questionable. particular, little is known about calibration, a property that allows confidence scores be interpreted as probabilities model's predictions true. Using large diverse European dataset, we investigate whether deep species classification in images well calibrated. Additionally, traps often configured take multiple photos same event, also explore calibration aggregated across sequences Finally, study effect practicality post‐hoc method, i.e. temperature scaling, made at image sequence levels. Based on five established three independent test sets, show averaging logits over sequence, selecting an appropriate architecture, optionally using scaling can produce well‐calibrated models. Our findings have clear implication for, instance, calculation error rates or selection score thresholds studies making use artificial intelligence

Language: Английский

Citations

3
Exploring five indicators for the quality of OpenStreetMap road networks: a case study of Québec, Canada DOI Creative Commons
Milad Moradi, Stéphane Roche, Mir Abolfazl Mostafavi

et al.

GEOMATICA, Journal Year: 2021, Volume and Issue: 75(4), P. 178 - 208

Published: Dec. 1, 2021

OpenStreetMap (OSM) is one of the most well-known volunteered geographic information (VGI) projects that aims to produce a free-world map. However, there are serious concerns about its quality. Numerous studies have assessed quality OSM by comparing database with reference database. Several researchers proposed use indicators as variables can describe in regions where no data available. A indicator variable has significant monotonic relationship measures. In this study, literature review was conducted identify and define main measures for assessing linear features. Owing limited access current data, only three elements—completeness, positional accuracy, attribute accuracy—were evaluated study. These were then used assess roads province Quebec. Finally, Spearman’s rank correlation coefficient test applied determine whether between related elements five potential indicators: population, average income, density roads, buildings, number points interest (POI). The contribution study testing following hypothesis: “There mentioned elements”. Statistical analysis showed terms completeness, population best indicators; income completeness indicator. All correlations quality, except two pairs (attribute roads) buildings). This proposes POI new not been found review.

Language: Английский

Citations

19