Smartphone imaging tool supports early detection of cancer

Dental Tribune International

Wed. 8 April 2026

save

HOUSTON, US: Because oral cancer is still frequently only diagnosed at an advanced stage, there is a need for adjunctive tools that support early detection and assist clinicians in referral decision-making. A recent study has evaluated a smartphone-based imaging system designed to aid dental professionals in this regard. The findings indicate that the system may support the early identification and triage of oral mucosal lesions, particularly in low-prevalence community dental clinic settings.

The smartphone system, called mDOC, captures paired white light and autofluorescence images of the oral cavity, enabling visualisation of tissue changes that may not be apparent under conventional lighting, and collects patient risk factors such as age, smoking history and lesion location. These inputs are analysed to generate a recommendation on whether specialist evaluation is warranted. The imaging process takes approximately 3.5 minutes, allowing integration into routine dental examinations with little disruption to workflow.

The study aimed to optimise and evaluate the mDOC system and its referral algorithm in a routine clinical setting. It was employed for the examination of 50 patients presenting for care at UTHealth Houston School of Dentistry and Harris Health Dental Center. The system’s referral recommendations were compared with those of a dental hygienist, a general dentist and an expert clinician.

According to the findings, the system could fairly accurately distinguish suspicious lesions from harmless ones. It detected more potentially dangerous lesions than the dentists and hygienists did, supporting its potential value as an adjunctive tool. However, the system showed only moderate sensitivity, correctly identifying 60% of lesions that experts recommended for referral. Additionally, it correctly identified most harmless lesions, but it produced a number of false positives, indicating the need for further optimisation.

“Early detection of oral cancer is essential because survival rates decrease as the disease progresses,” lead author Dr Ruchika Mitbander, a postdoctoral fellow at Rice University Department of Bioengineering in Houston, said in a press release. “Yet, in community dental settings—especially in underserved or low-resource regions—providers often lack tools to reliably identify lesions that require specialist referral.”

Notably, unaided clinicians in the study did not identify any of the cases requiring referral, suggesting that adjunctive tools may enhance detection performance in routine practice. However, the moderate sensitivity also highlights that such systems should complement, not replace, clinical judgement.

According to the authors, compared with existing approaches, mDOC offers several advantages. It combines dual-mode imaging in a portable format and integrates multiple data inputs into a single decision support system. Dr Mitbander suggested that, beyond screening, the system may support monitoring of lesions and could be combined with additional diagnostic methods, such as cytology, particularly in settings with limited specialist access.

“The mDOC system has the potential to play a significant role in the early detection of oral cancer in dental care settings,” Dr Mitbander concluded, citing improved patient outcomes and amelioration of disparities in access to specialist care, particularly in underserved or rural communities.

Research has demonstrated that dentists and dental hygienists can play a meaningful role in the detection of precancerous oral lesions and oral cancers. However, training and experience in reliably distinguishing suspicious lesions from benign conditions are not uniform across settings, and some providers may have limited preparation in this regard. Therefore, decision support tools may help bridge the gap between routine screening and specialist referral in general dental practice.

The study, titled “Optimization of a mobile imaging system to aid in evaluating patients with oral lesions in a dental care setting”, was published online on 6 October 2025 in Biophotonics Discovery.