Detecting Trichiasis Using 3D Photography: A Case Study

This blog first appeared in Medium on February 12, 2019

Trachoma, caused by the bacterium Chlamydia trachomatis, is the leading infectious cause blindness, resulting in visual impairment in 1.9 million people globally. While it has been eliminated from many countries, it continues to be a major public health concern in 37 more. Ethiopia has the highest Trachoma burden; around 69 million people live in endemic regions requiring control interventions.

Infection with Chlamydia trachomatis causes chronic conjunctivitis. If repeated infections occur, scarring builds up in the surface membranes of the eye, resulting in the inturning of eye lashes known as trachomatous trichiasis (TT). The rubbing of the eyelashes against the cornea causes painful blinking, excessive tearing, and eventual vision loss, if untreated.

Trachoma is being controlled through the application of the SAFE Strategy, which stands for “S” surgery for trichiasis, “A” antibiotics, “F” facial cleanliness and “E” environmental improvements. Trachoma is often concentrated in low-resource areas where sanitation is limited. It spreads from contact with the infected individual’s ocular discharge: through personal or contaminated surface contact, flies, clothing, or towels that have traces of infection.

If people who develop inward-directed eye lashes are identified early on, the surgery usually prevents sight loss. Early detection of TT is important to allow for timely intervention, which involves examinations by trained disease graders. However, in settings where the prevalence of TT is low, it is difficult to find enough patients to take part in grader training exercises so that graders become proficient. Therefore, we need alternatives to train graders.

In a study funded by UK aid from the British people, researchers developed and then evaluated a methodology using 3D images to teach and test trachoma graders-in-training. Compared to 2D photos, the approach using 3D images closely resembles actual examination. The study team included investigators from Amhara Regional Health Bureau, The Carter Center, and the International Centre for Eye Health at the London School of Hygiene & Tropical Medicine (LSHTM).

“Field assessments have always been the gold standard for diagnosing trichiasis,” said Matthew Burton, principal investigator of the study and Professor of International Eye Health at LSHTM. “However, as countries approach the trachoma elimination targets, there will be relatively few people available to train health workers on clinical signs. A suitable alternative training approach is needed. 3D photographs may provide this option to reliably train people to recognize this condition.”

Field graders assess TT by counting number of lashes coming in contact with the eye. Findings from a previous study indicated that in comparison to field assessment, 2D images limit the perspective of the patient’s eyelashes, thereby increasing the likelihood for over-identifying TT in the patients pictured. 3D imaging, on the other hand, allows the examiner to judge depth and determine if the lashes are touching the eye, so relying on this training tool can provide an opportunity to enhance the quality of training.

To determine the value of 3D imagery during training, the researchers first assessed how results from 2D and 3D image grading compared to live grading of the same TT patient. Then, a group was trained using 3D images to see how their diagnostic accuracy compared with that of experienced trichiasis graders.

3D TT Study Picture


The study population, consisting of 26 health professionals, took part in four days of training sessions on TT so researchers could determine the reliability of 3D image training for field grading preparation. The sessions covered a variety of topics, such as the anatomy of the eye, eye care practices, eye examination, blindness, trachoma, TT, and more. Trainees were then presented with 2D images of trachoma and TT cases on a screen, followed by an explanation on how to recognize TT case presentations in 3D images. Their final assessment involved determining the patients with and without trichiasis in a set of 3D pictures, with 30-second limit per eye image. After completing the sessions and assessment, trainees went into the field to do the same identification on live patients, and their performance was compared to that of an expert trachoma grader. The final part of the performance evaluation compared the accuracy of case identification of 3D picture-trained individuals to that of the integrated eye care workers.

The findings of this study demonstrated the following: trainees were notably better at determining the absence of trichiasis with the 3D images compared to the 2D images. They identified presence or absence of TT at a similar level to the experienced health workers. Interestingly, trainees performed better during live field grading than they did when looking at 3D images, suggesting that it may be more difficult to identify TT in the 3D photographs. However, this outcome may have been influenced by their 3D test feedback prior to their live case identification. Evaluation forms revealed that around 80% of trainees felt that image-based training was easy to do and adequately prepared them for the real-life patient grading.

While earlier studies examined face-to-face trachoma grading, this study provides insight to the role 3D photography has in case identification, training methods, and clinical performance evaluation for trachoma. It can potentially inform other disease diagnostic approaches that similarly require a closer look to enhance diagnostic skills. Although the disease progresses slowly, the consequences of late or undiagnosed trachoma are severe, so identifying cases is essential to prevent unnecessary suffering. Ultimately, integrating 3D image-based training can promote confidence in the recognition of trachoma and improve quality of training by showing trainees various clinical presentations. This advance will help move trachoma elimination programs a step closer to a future free from TT.


1. Children in Ethiopia. Trachoma, a bacterial infection, can be spread by flies. Photo by Chelsea Toledo, NTD Support Center.

2. Trichiasis graders (a) practicing grading trichiasis using 3D images, and (b) performing the intergrader 3D image test. Image Courtesy of Matthew Burton.