Thermal Imaging & AI for Onchocerciasis

 Have you ever heard of Onchocerciasis? Did you know that it is responsible for blindness in over half a million people worldwide? In this blog, let's take a deeper dive into Onchocerciasis, its impact on affected communities, the challenges associated with detection, and the innovative use of Thermal Imaging technology in combating this disease.

What is Onchocerciasis?

Onchocerciasis, commonly known as river blindness, is a tropical disease caused by the filarial worm called Onchocerca volvulus. This disease primarily affects millions of people residing in sub-Saharan Africa and parts of South America, particularly in rural and impoverished communities. The transmission of Onchocerciasis occurs through the bites of infected blackflies of the genus Simulium, which breed in fast-flowing rivers and streams. 

Once inside the human body, the larvae of Onchocerca volvulus undergo a complex development process, maturing into adult worms within approximately 12 to 18 months. The adult worms can be male and female and can attain considerable sizes. Male worms typically grow to lengths ranging from 1.5 to 4.5 cm, while their female counterparts can reach impressive lengths of up to 50 cm.  

Most adult female worms form fibrous nodules under the skin and sometimes near muscles and joints. These nodules also protect them from the human immune response. Adult male worms are usually found near the female worms. The female adult worms produce thousands of new larvae each day, perpetuating the cycle of infection within the human body. The larvae become detectable in the skin 12–18 months only after the initial infection. Over time, the presence of Onchocerca volvulus within the body can lead to a range of debilitating symptoms. Skin lesions, severe itching, and visual impairment are common manifestations of the disease, with the potential for irreversible blindness if left untreated. The skin lesions and itching are often so severe that affected individuals may experience significant discomfort and disability, impacting their quality of life and ability to perform daily tasks.



Challenges in Detecting Onchocerciasis

Despite continuous efforts to control and eradicate onchocerciasis, significant challenges persist, particularly in remote regions with limited access to healthcare resources. While Ivermectin administration serves as a crucial preventive measure, it targets only the microfilariae (eggs) of the Onchocerca volvus, leaving the adult worms unaffected. To effectively eliminate the disease, it becomes imperative to identify and remove the adult worms residing within the body, often necessitating surgical intervention such as nodulectomy to excise the nodules formed by the female worms.

Detecting onchocerciasis presents formidable obstacles. Conventional diagnostic techniques, including skin snips and serological tests, are often time-consuming, labor-intensive, and reliant on specialized equipment and trained personnel. Moreover, the clinical manifestations of onchocerciasis, such as skin lesions and visual impairment, can closely mimic those of other dermatological conditions and eye diseases, further complicating accurate diagnosis. There is a pressing need for innovative and accessible diagnostic tools that can overcome the limitations of traditional methods and facilitate timely and accurate detection of onchocerciasis.

Thermal Imaging for Onchocerciasis Detection

The female onchocerca worms generates more heat due to several factors inherent to the pathogenesis of onchocerciasis. The core of onchocerca nodules is a dense infiltrate of inflammatory cells in which microfilariae are released, a contributing factor to localized heat generation. Further, it is also found that the formation of nodules is associated with angiogenesis (formation of new blood vessels). This can further influence the heat distribution in the affected tissue and enhances the thermal signature. And lastly, majority of these nodules typically resides in subcutaneous tissue (close to skin surface), facilitating easier heat transfer to the surface. 

As we discussed in our previous blog posts, the modern thermal cameras can discern minute temperatures up to 50 mK (0.05°C).  This allows one to study the subtle thermal variations induced by onchocerca nodules by capturing their thermal images. However, these variations might be difficult to visualize with naked eye and there are no existing visual interpretation protocols for onchocerciasis detection. In this context, recent studies have demonstrated the promising potential of artificial intelligence (AI) algorithms in deciphering the thermal variations induced by onchocerca nodules. 

AI + Thermal Imaging for Onchocerciasis Detection

Broadly, the use of AI for analysis of thermal images to detect onchocerciasis can be seen in three fronts:
  • Triage individuals for Onchocerciasis Detection: The initial step in combating onchocerciasis involves the timely identification and triaging of individuals requiring treatment. AI algorithms, when coupled with thermal imaging technology, facilitate the systematic capture and analysis of thermal images from various body parts. Through machine learning techniques, these algorithms can discern subtle thermal variations characteristic of Onchocerca nodules, thereby enabling the classification of thermal images based on disease indication. By isolating individuals exhibiting thermal signatures suggestive of onchocerciasis, healthcare practitioners can prioritize treatment allocation and intervention strategies. 
  • Localization of Onchocerca Nodules: Surgical excision of Onchocerca nodules using nodulectomy is key in the management of onchocerciasis-related complications. AI-powered thermal image analysis can help in identifying the precise localization of these nodules within the body. This localization information guides surgical planning, enabling healthcare practitioners to target and excise Onchocerca nodules.
  • Fecundity Estimation of Onchocerca Nodules: Fecundity represents the reproductive status of the onchocerca worms. Reproductive oncho worms pose more threat compared to non-reproductive worms as they can release thousands of new larvae each day. The reproductive worms typically have higher and distinct heat pattern when compared to non-reproductive worms due to their high metabolic activity. AI-based thermal image analysis can offer a non-invasive means of assessing the reproductive status of these parasitic worms. This would be crucial in drug testing for understanding the treatment responses of different drugs. 

Open Source Datasets

If you want try out AI algorithms for the above problem statements, there is an open source dataset called Niramai Oncho Dataset from Niramai Health Analytix. This dataset consists of thermal images and videos of different body parts consisting of palpable nodules along with their histopathological statuses confirming the presence of female adult worm.

Overall, the dataset comprised thermal data (images/videos) from 125 participants captured for different body parts. In total, the dataset encompasses thermal data for 192 distinct body parts with 101 corresponding to live female nodules and the remain 91 corresponding to dead nodules.

This dataset can be leveraged  to develop AI-driven solutions tailored to address the above discussed problem statements in onchocerciasis detection and management. By employing machine learning techniques such as convolutional neural networks (CNNs) for image classification and segmentation, researchers can train models to accurately classify thermal images based on the presence of live female nodules, localize nodules within thermal images, and even estimate the reproductive status of the parasites.


References

[1] Dedhiya, R., Kakileti, S.T., Deepu, G., Gopinath, K., Opoku, N., King, C. and Manjunath, G., 2022, July. Evaluation of Non-Invasive Thermal Imaging for Detection of Viability of Onchocerciasis Worms. In 2022 44th Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC) (pp. 3518-3521). IEEE.
[2] Dedhiya R, Kakileti ST, Gopinath K, Edem A, Donkor B, Seidu AM, Attah SK, King CL, Opoku N, Manjunath G. Non-invasive Thermal Imaging for Estimation of the Fecundity of Live Female Onchocerca Worms. InMICCAI Workshop on Medical Image Assisted Blomarkers' Discovery 2022 Sep 18 (pp. 102-110). Cham: Springer Nature Switzerland.

Comments