Kolkata, India – Researchers have developed a low-cost, artificial intelligence-powered model capable of screening cervical cancer samples in as little as 30 seconds, offering a potentially transformative tool for early detection, particularly in resource-constrained settings. The study, conducted by a collaborative team from four Kolkata institutes, demonstrated high diagnostic accuracy and was recently published in the Asian Pacific Journal of Cancer Prevention.
Cervical cancer remains a significant global health challenge, but early detection through screening programs like the Pap smear – also known as the Pap test – can dramatically improve outcomes. The conventional Pap smear involves examining cells collected from the cervix under a microscope to identify precancerous or cancerous lesions. While the actual collection of the sample is relatively quick, the analysis in a laboratory can be time-consuming and costly.
“This AI-based cervical screening is very promising, especially in rural settings that see higher cervical cancer prevalence but lack of diagnostic facilities,” said Arindam Karmakar, an associate professor of pathology who contributed to the study. The new AI model aims to address these limitations by automating the analysis process and reducing costs.
How the AI Model Works
The researchers developed AI programming software designed to automatically detect and classify cells from microscopic images of Pap smear slides. These images were captured using standard Android phones or tablets, making the technology readily accessible and affordable. The system analyzes the cell morphology – the shape and structure of the cells – to diagnose potential abnormalities.
The study involved analyzing 292 smear samples obtained from hospital-based labs at Calcutta National Medical College (CNMC) and NRS Medical College. The Pap smear images were rigorously analyzed using the newly developed AI software, adhering to standardized validation methods.
Promising Accuracy Rates
The results showed the AI model successfully classified 98.1% of normal cells and 80.5% of abnormal cells within the hospital samples. Importantly, the team also found a significant correlation between the AI’s findings and results from a biopsy – considered the gold standard for confirming a cervical cancer diagnosis.
The potential cost savings are substantial. Currently, a Pap smear in Kolkata averages around Rs 1,000 (approximately $12 USD). Researchers estimate that utilizing this AI software could reduce the screening cost to as low as Rs 200 (approximately $2.40 USD), significantly increasing accessibility for women who might otherwise be unable to afford regular screening.
The Broader Context of Cervical Cancer Screening
Advances in cervical cancer screening and prevention have been ongoing for decades. While Pap smears have been used since the 1940s, the discovery that the human papillomavirus (HPV) causes more than 90 percent of cervical cancers has led to more accurate screening tests and, crucially, the development of an HPV vaccine, approved by the U.S. Food and Drug Administration in 2006.
Despite these advancements, challenges remain. According to research, cervical cancer continues to claim the lives of approximately 4,000 women in the United States each year. Effective screening programs are vital, but access and affordability remain barriers for many, particularly in low-resource settings.
AI’s Role in Improving Outreach and Accuracy
The study’s findings suggest that AI-driven tools have the potential to revolutionize cervical cancer screening by improving outreach, availability, accuracy, and cost-effectiveness. The speed of the AI analysis – less than 30 seconds per sample – is a significant advantage, allowing for higher throughput and faster results.
Priyabrata Das, scientific director at CliniMed Lifesciences and a member of the research team, emphasized the speed and accuracy of the AI tool. “Screening using this AI tool took less than 30 seconds and the accuracy rate was high,” he stated.
Responsible Implementation is Key
Experts caution that responsible and ethical implementation, along with continued validation, are crucial for maximizing the benefits of this technology. “If implemented responsibly, ethically, and with continued validation, such AI-driven tools can significantly accelerate India’s progress towards reducing cervical cancer mortality,” said Snehendu Konar, clinical operations lead at Auriga Research.
The research team included contributions from Arindam Ray, Soumyabroto Banerjee, Swati Dasgupta, Sayan Dey, Umar Faruk, Moumita Maiti, and Prosenjit Saha, highlighting a collaborative effort to address a critical public health need. The development of this AI model represents a significant step forward in making cervical cancer screening more accessible and affordable, potentially saving lives and reducing the burden of this disease worldwide.
The study, titled ‘Evaluation of the Diagnostic Accuracy of Cervical Cell Morphologies from Android Device-Captured Cytopathological Microscopic Images through Artificial Intelligence in Mainly Rural or Resource-Constraint Areas of India’, underscores the growing role of artificial intelligence in transforming healthcare delivery, particularly in areas where resources are limited.
