Source TGMalaria Topic Driver Title Att3 Overview of the topic area TGMalaria Contact Rose Nakasi Email gnakasirosegmailcom Abstract This PPT contains a presentation on ID: 1010795
Download Presentation The PPT/PDF document "FGAI4H-R-014-A03 Cambridge, 21-24 March ..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
1. FGAI4H-R-014-A03Cambridge, 21-24 March 2023Source:TG-Malaria Topic DriverTitle:Att.3 - Overview of the topic area (TG-Malaria)Contact:Rose NakasiE-mail: g.nakasi.rose@gmail.com Abstract:This PPT contains a presentation on TG-Malaria.
2. Topic Group-Malaria: AI based detection of Malaria-an updateMeeting RRose Nakasi, PhDg.nakasirose@gmail.comMakerere University, UgandaE-meeting, 21st – 24th March 2023
3. Malaria burden in endemic CountriesAccounts for over 3.4 billion cases globally Lack of enough trained lab technicians1.72 microscopes per 100,000 population, but only 0.85 Microscopists per 100,000Gold standard diagnosis(microscopy) challengeSOP requires not to view more than 30 slides a dayLess diagnosis throughputVariations in individual expert judgmentAI solutionSupports image analysis and has potential to improve the timeliness and accuracyThere is need to;Standardise benchmarking AI solutions for the detection of MalariaBackground
4. Digital imagingSetup
5. Digital Microscopy imaging analysisInference using CNN model
6. Quality datasets neededHave more datasets for training and testingWell labelled datasets SolutionAI models and approaches related to malaria detection.Suggestions on scoring metrics.Improvements on the benchmarking framework.Support to the group on different aspects (data, methods, benchmarking, etc.) of this topicExtension of the solution to improve disease surveillance and prediction. Heterogeneous Data neededTG-Malaria activities
7. Updates since meeting QOnline meetingsMigrating competion to Codabench https://www.codabench.org/competitions/389/?secret_key=2f16646d-f5db-40c1-ae57-6a89c6802551focusing on benchmarking instead of competitionImplementing classification task on codabenchDiscussions on the outcomes of the implemented platform
8. User interface for the CodabenchNew Benchmarking-Malaria platform with codabench
9. Current Tasks undertakenDataset curation based on standardised data collection protocolDeveloping a standardised data collection protocolPublic dataset under collectionThick Blood smear imagesThin blood smear image
10. Iterating with different algorithms to assess performance on new datasets collectedFor tasks like parasite detectionparasitemia determinationspecies identification growth stage identification, and artifact detection.Creating a new challenge to be tested on our developed codalab benchmarking platform using the new benchmarking dataset to a wider communityPublishing outcomes of analyticsNext steps
11. Call for participationParticipation can be in form of:Provision of quality labelled dataAI models and algorithms for benchmarking task on malariaGeneral support on different aspects of this topic (data, methods, benchmarking, etc.)11
12. Contact usTG-Malariafgai4htgmalaria@lists.itu.intTG-Driverg.nakasirose@gmail.com
13. Topic Group members:NameAffiliationPhilippe VerstraeteCo-founder of Milan and Associates, ItalyLaura Moro, Researcher, science & medical writer. Co-founder of AI Scope, SpainDr. Helmi Zakariah. Cofounder of AIME company, MalaysiaMartha Shaka Researcher at University of Dodoma, TanzaniaAna Rivière Cinnamond Advisor and Pubic Health Expert under Health Emergency Information & Risk Assessment Department with PAHO/WHO.Rigveda Kadam Senior Access Officer, FIND, SwitetzerlandSeda Yerlikaya Scientific Officer, FIND, SwitetzerlandHerilalaina RAKOTOARISON PhD student in Machine learning from the Université Paris-Saclay) AI for Outbreak Detection - FG-AI4H13