Page 45 - DIGITAL e-Book RCPH 2026

Organised by:
1 REGIONAL CONFERENCE onon O r g a n i s e d b y :
E
E
I
G
C
E
N
R
E
C

N
O
L
F
O
A
N

R

1

st st
E

P P R E C I S I O N H E A L T H
I
C
PRECISION HEALTH
H
I
H

N
O
E
R
T
L
A
S
Abstracts for 1st Regional Conference on Precision Health (RCPH)
15-16th April 2026, Royale Chulan Kuala Lumpur
From Apps to Access: Digital Triage and Community Pathways for Early
Detection of Oral Cancer
Professor Dr Sok Ching Cheong
Cancer Research Malaysia
ABSTRACT
Oral cancer remains a major public health challenge, particularly in low- and middle-income settings
where late-stage diagnosis is common and access to specialist care is limited. In Malaysia and across
Asia, a substantial proportion of patients present at advanced stages, resulting in significantly poorer
survival outcomes. Early detection is therefore critical, yet conventional screening models are
constrained by centralised services, workforce limitations, and geographical barriers. I will discuss our
work on MeMoSA® (Mobile Mouth Screening Anywhere), a digital health platform developed to enable
community-based oral cancer screening through mobile technology, teleconsultation, and artificial
intelligence (AI). By leveraging the widespread availability of smartphones and the visual nature of oral
lesions, MeMoSA enables non-specialist personnel to capture structured clinical images and risk
information, which are then reviewed by clinicians. The platform integrates multiple components,
including AI-guided image capture, digital data collection, and clinician-facing reporting systems,
facilitating a scalable approach to early detection. Real-world implementation demonstrates that such a
model can extend screening reach, improve access to care, and identify individuals with lesions at risk
of malignancy who may otherwise remain undetected. A key component of this approach is the
development of AI models trained on a large, multi-country dataset of annotated oral images, enabling
risk stratification and referral decision support. These models show promising performance in
identifying high-risk lesions and support more efficient use of specialist resources. Importantly,
MeMoSA represents a shift from specialist-centred screening towards community-embedded, digitally
enabled pathways, aligning with broader precision health goals of delivering the right care to the right
population at the right time. This work demonstrates that integrating digital tools, AI, and community-
based implementation can help close critical gaps in early cancer detection, offering a scalable model
for improving equity and outcomes in underserved populations.