Submitted
2025 Global Health Challenge

Octopi

Team Leader
Pranav Shrestha
We propose a platform technology, Octopi, which combines automated microscopy with machine learning to enable accurate, scalable and quantitative detection of pathogens and pathological conditions. For detecting malaria parasites, Octopi’s AI algorithms take advantage of a unique spectral shift in the parasites stained with a nuclear stain, DAPI, and imaged using fluorescence microscopy. Our approach enables i) processing up to...
What is the name of your organization?
Cephla Inc.
What is the name of your solution?
Octopi
Provide a one-line summary or tagline for your solution.
AI-powered automated microscopy for accurate, scalable and quantitative detection of malaria and other diseases
In what city, town, or region is your solution team headquartered?
Mountain View
In what country is your solution team headquartered?
USA
What type of organization is your solution team?
For-profit, including B-Corp or similar models
Film your elevator pitch.
What specific problem are you solving?
Microscopy remains a foundational pillar and a WHO gold standard for diagnosing multiple diseases worldwide, including infectious diseases like malaria and tuberculosis, and non-communicable diseases like sickle cell disease and cancer. Despite being ubiquitous due to its utility in a variety of healthcare settings, including remote and low-resource settings, manual microscopy requires highly trained personnel, is time consuming and is error prone. Recent advances in artificial intelligence (AI) and robotics offer opportunities to automate and enhance microscopy, enabling high-throughput and multi-disease diagnostics. However, current automated microscopy platforms are often costly and inflexible. Manual microscopy is one of the gold standard techniques to detect malaria but is associated with low sensitivity, especially at low parasite densities and for asymptomatic cases that are reservoirs for infection. Malaria still remains a major global health problem, with over 260 million cases and around 600,000 deaths annually. Other tests have their own limitations: rapid diagnostic tests (RDTs), while widely used, can miss infections with emerging gene deletions, cannot quantify parasite density for severity assessment and treatment monitoring, and have persistent positivity after parasite clearance. Thus, there is a need for accurate, scalable and quantitative diagnostic tools to address these critical gaps.
What is your solution?
We propose a platform technology, Octopi, which combines automated microscopy with machine learning to enable accurate, scalable and quantitative detection of pathogens and pathological conditions. For detecting malaria parasites, Octopi’s AI algorithms take advantage of a unique spectral shift in the parasites stained with a nuclear stain, DAPI, and imaged using fluorescence microscopy. Our approach enables i) processing up to 1 million blood cells per minute, which is orders of magnitude faster than current techniques; ii) high sensitivity and specificity (>97% in preliminary clinical studies); iii) low limit of detection (~12 parasites/μL, outperforming standard methods); and iv) quantitative measurements essential for treatment monitoring. The platform uses standard glass slides and widely available consumables for integrating easily into current healthcare systems and for reducing the need for expensive proprietary consumables. Its modular, open-source design allows adaptation for detecting multiple diseases beyond malaria, including tuberculosis and sickle cell disease. Octopi's cloud infrastructure enables continuous AI improvement and data management, creating what we call an "App Store" for microscopy-based diagnostics, where researchers can develop new disease detection applications on our unified platform. Video demo for real-time detection of malaria Plasmodium falciparum parasites from clinical samples: https://youtu.be/SYr0SZw7Fow
Who does your solution serve, and in what ways will the solution impact their lives?
Octopi aims to directly serve populations in malaria-endemic regions, particularly within low- and middle-income countries and resource-constrained settings where diagnostic capacity is often lowest despite the highest disease burden. Young children in sub-Saharan Africa are disproportionately affected and largely underserved - in 2023, less than half of febrile children under five in the WHO African Region received malaria testing, despite accounting for 74% of the deaths in the region. These communities lack sufficient trained personnel, infrastructure, and reliable diagnostic tools. Octopi directly impacts the lives of the patients and those of the healthcare professionals by bridging critical diagnostic gaps in the current system. Automation and AI-powered disease detection reduces the need for specialized training, while Octopi’s portability and use of standard glass slides address infrastructure limits. By offering rapid and highly sensitive detection and quantification, Octopi will enable earlier diagnosis, effective treatment monitoring, and identification of asymptomatic carriers, crucial steps for disease control. Octopi aims to empower local healthcare workers with an accurate, easy-to-use tool, ultimately improving health outcomes and strengthening surveillance in the most vulnerable communities.
Solution Team:
Pranav Shrestha
Pranav Shrestha
Clinical Research Scientist