Hezerul Abdul Karim

Dr. Hezerul Abdul Karim

Multimedia University, Cyberjaya Campus, Malaysia.
orcid

Research Focus & Metrics

Academic Career Length

23Years (2004-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
7 May 2025 in Frontiers in Digital Health
Verified
Journal Article
Open access
PDF file
VFVasco Gerardo Hinostroza Fuentes
Vasco Gerardo Hinostroza Fuentes
Hezerul Abdul KarimHezerul Abdul Karim
Hezerul Abdul Karim
MTMyles Joshua Toledo Tan
Myles Joshua Toledo Tan
NANouar AlDahoul
Nouar AlDahoul
The healthcare industry continues to face significant operational challenges in patient care, resource allocation, and administrative processes. For instance, despite spending 16.8% of its gross domestic product on healthcare by 2015, the United States reported higher rates of preventable hospitalizations and lower life expectancy compared to countries that spent nearly half as much (Papanicolas et al., 2019). In fact, the average life expectancy in the United States was 78.8 years, falling short of the 80.6-year average among OECD countries (Papanicolas et al., 2019). Moreover, 73.2% of insured adults reported experiencing at least one administrative burden that led them to delay or forgo medical care (Kyle & Frakt, 2021). These inefficiencies stem not only from financial concerns but also from deeply embedded flaws in the administrative and technological infrastructure of the healthcare system. Such persistent inefficiencies highlight a need not merely for automation, but for intelligent, adaptive systems capable of navigating complexity in real time.A major driver of these challenges is administrative overhead. Healthcare institutions allocate approximately 20% of their budgets to administrative tasks, while American physicians spend around 13% of their work time on similar responsibilities (Cutler & Ly, 2011). Compounding this issue are fragmented workflows, excessive documentation, and poorly integrated clinical systems, which increase physician burnout and the likelihood of clinical errors (Zhang & Padman, 2014). For example, some computerized provider order entry systems are not tailored to patient needs, requiring 10% more physical effort than manual order selection (Zhang & Padman, 2014). Additionally, order sets can become outdated quickly, reducing their clinical value. These challenges underscore the need for a more intelligent system to ease administrative burdens and support effective clinical decision-making. What is needed is not just an AI system that follows static rules, but one that learns, evolves, and operates with autonomy.Agentic artificial intelligence (AI) offers a promising solution by autonomously managing complex healthcare tasks, reducing human error, and enhancing efficiency (Acharya et al., 2025). Using machine learning (ML) algorithms, agentic AI adapts to real-time healthcare environments (Jiang et al., 2017).Unlike conventional AI, which depends on fixed rules, agentic AI acts on its own to achieve healthcare goals and continuously updates its behavior as new information comes in. It can streamline workflows, enhance diagnostic accuracy, and reduce administrative workload (Jiang et al., 2017;Wubineh et al., 2014). Some agentic AI systems have been shown to lower cognitive workload by up to 52% (Zhang & Padman, 2014). Predictive models powered by agentic AI can identify patients at risk of disease progression or complications, resulting in fewer hospitalizations, reduced healthcare costs, and better outcomes (Khaleelullah et al., 2024). For instance, AI-based monitoring systems can detect subtle changes in vital signs, predict deterioration, and alert clinicians before critical issues develop, enabling timely intervention.Because agentic AI is goal-driven and adapts over time, it is especially well-suited to handle the complexity of hospital environments and ever-changing patient needs. In addition, agentic AI can optimize hospital resource management by dynamically adjusting staffing, supply distribution, and patient flow based on real-time data (Acharya et al., 2025). This perspective introduces an agentic AI framework designed to automate, optimize, and personalize medical services. Unlike traditional ML models, agentic AI continuously learns from routine data and adjusts its responses to match evolving healthcare demands. This combination of adaptiveness and autonomy makes agentic AI not just an innovation, but a necessity for the future of healthcare delivery. By reducing administrative burdens, enhancing clinical decision-making, and streamlining operations, agentic AI has the potential to transform healthcare delivery and improve both outcomes and cost-effectiveness.Reclaiming Time for Care: The Potential Administrative Power of Agentic AI AI enhances administrative efficiency by automating routine tasks such as medical documentation, insurance claims processing, patient scheduling, and staff coordination (Bekbolatova et al., 2024;Knight et al., 2023). In the United States, healthcare professionals spend approximately 25% of their working hours on administrative duties (Bekbolatova et al., 2024). This administrative burden contributes to physician burnout, which in turn affects the quality of care delivered to patients (Esmaeilzadeh, 2024). Automating repetitive and time-consuming administrative tasks can therefore free up valuable time for medical professionals to focus more on direct patient care. Agentic AI builds upon this by not only executing administrative tasks but also proactively refining and restructuring workflows. AI-driven automation addresses this issue by significantly reducing manual workloads, enabling healthcare providers to allocate their time more intelligently and efficiently (Alowais et al., 2023;Liu et al., 2023).One of the most time-consuming administrative tasks is the documentation of clinical history. Physicians often spend more than an hour documenting electronic health records (EHRs) for every hour spent with patients, contributing to both burnout and inefficiencies (Reddy, 2024). To address this, AI-powered natural language processing and voice recognition technologies are being integrated into EHR systems to assist with medical transcription (Sahithya et al., 2024). These tools allow physicians to dictate patient notes, which are then structured into standardized documents by AI. When these systems use agentic AI, they can learn each clinician's preferences over time to improve documentation quality. This...

2 May 2024 in IEEE Access
Verified
Journal Article
Open access
ZKZu Jun Khow
Zu Jun Khow
YTYi Fei Tan
Yi Fei Tan
Hezerul Abdul KarimHezerul Abdul Karim
Hezerul Abdul Karim
HAHairul Azhar Abdul-Rashid
Hairul Azhar Abdul-Rashid
The rapid advancements in deep learning have revolutionized the field of computer vision. However, despite the significant progress in computer vision, there remains a scarcity of research focused on utilizing this technology for distance estimation. Exploring such studies can bring immense convenience to people, especially in applications like anomaly object detection. On that account, this research proposes an improved detection model based on You Only Look Once version 8 (YOLOv8) namely YOLOv8-CAW, which is capable of both detecting target objects and accurately calculating their distances. The proposed method involves incorporating the Coordinate Attention and Wise-IoU into the YOLOv8 network, enhancing the detection accuracy. Combined with the distance estimation algorithm, results in a comprehensive output that includes both detection results and calculated distances. At the end of the experiment, a substantial improvement in performance metrics was observed, the model achieved increases in recall (0.4%), precision (2.2%), and Mean Average Precision (mAP) (1.5%) within the 0.5 to 0.95 threshold range, while maintaining inference speeds similar to the baseline model in PASCAL VOC dataset. Besides that, distance estimation achieved an approximate average accuracy of 90% which shows the results are highly encouraging and promising. The successful integration of computer vision and distance estimation opens new possibilities for practical applications, showcasing the potential of this approach in real-world scenarios.

7 January 2025 in Frontiers in Public Health
Verified
Journal Article
Open access
PDF file
MTMyles Joshua Toledo Tan
Myles Joshua Toledo Tan
HKHarishwar Reddy Kasireddy
Harishwar Reddy Kasireddy
ASAlfredo Bayu Satriya
Alfredo Bayu Satriya
Hezerul Abdul KarimHezerul Abdul Karim
Hezerul Abdul Karim
NANouar AlDahoul
Nouar AlDahoul
Hyper-personalized medicine represents the cutting edge of healthcare, which aims to tailor treatment and prevention strategies uniquely to each individual. Unlike traditional approaches, which often adopt a one-size-fits-all or even broadly personalized approach based on broad genetic categories, hyper-personalized medicine considers an individual’s comprehensive health data by integrating unique biological, genetic, lifestyle, and environmental influences. This method goes beyond simple genetic profiling by recognizing that health outcomes are influenced by complex interactions among our environment, daily routines, and physiological processes and responses.Central to hyper-personalized medicine is the integration of lifestyle and environmental factors. Lifestyle habits, such as diet (Dalwood et al., 2020; Genel et al., 2020; Marx et al., 2020; Hepsomali & Groeger, 2021; Dinu et al., 2022; Yang et al., 2022; Sadler et al., 2024), exercise (Chow et al., 2022; Qiu et al., 2022; Ross et al., 2022; D’Onofrio et al., 2023; Isath et al., 2023; Mahindru et al., 2023; Ashcroft et al., 2024; Ponzano et al., 2024), and sleep patterns (Hepsomali & Groeger, 2021; Baranwal et al., 2023; Eshera et al., 2023; Lim et al., 2023; Sletten et al., 2023; Uccella, 2023; Weinberger et al., 2023), directly impact health. Hence, understanding these factors helps tailor interventions that align with the day-to-day realities of an individual. Environmental factors, such as air quality (Cheek et al., 2020; Markandeya et al., 2020; Shukla et al., 2022; Tang et al., 2022; Abdul-Rahman et al., 2024; Bedi & Bhattacharya, 2024), climate (Coates et al., 2020; Ebi et al., 2021; Helldén et al., 2021; Reismann et al., 2021; Rocque et al., 2021; Zhang et al., 2021; Münzel et al., 2024; Palmeiro-Silva et al., 2024), and exposure to pollutants (Qadri & Faiq, 2019; Petroni et al., 2020; Lin et al., 2022; Sun et al., 2022; Xu et al., 2022; Yu et al., 2022; Levin et al., 2023; Shetty et al., 2023; Deziel & Villanueva 2024; Sharma et al., 2024), also play significant roles in determining health outcomes. By continuously monitoring and analyzing these elements, healthcare providers can create dynamic health plans that adapt to real-time changes. This would allow for proactive measures and optimized care.To enable such a complex model of care, advanced technologies like quantum computing, artificial general intelligence (AGI), internet of things (IoT), and 6G connectivity play crucial roles. Quantum computing offers the ability to process vast and intricate datasets, such as those required to model interactions between genetic markers, environmental exposures, and lifestyle choices, with far greater speed and accuracy than classical computing (Munshi et al., 2023; Kumar et al., 2024; Stefano, 2024; Ullah & Garcia-Zapirain, 2024; Yu et al., 2024). AGI, with its adaptive learning capabilities, can analyze and make sense of this data to provide precise, evolving recommendations that change as a patient’s environment or lifestyle does (Liu et al., 2024; Mitchell, 2024; Sun et al., 2024; Tu et al., 2024). IoT devices, including wearables and environmental sensors, gather continuous data from individuals, tracking physical activity, biometrics, and environmental conditions like air quality and humidity (Puri et al., 2021; Islam et al., 2024; Mathkor et al., 2024; Rocha et al., 2024; Šajnović et al., 2024; Salam, 2024). With the advent of 6G connectivity, this data is seamlessly transferred and processed in real time, enabling instant feedback and intervention (Nayak & Patgiri, 2021; Nguyen et al., 2021; Ahad et al., 2024; Kumar, Kaur, et al., 2024; Mahmood et al., 2024; Mihovska et al., 2024).Together, these technologies form the backbone of a hyper-personalized healthcare model, which will push beyond traditional medical practices to create a highly responsive, individual-centered approach to health. As these advancements continue to evolve, hyper-personalized medicine has the potential to fundamentally reshape healthcare, offering truly personalized interventions that support long-term health and well-being.

9 June 2025 in Digital
Verified
Journal Article
Open access
PDF file
Mosharof Hossain DipoMosharof Hossain Dipo
Mosharof Hossain Dipo
Fahmid Al FaridFahmid Al Farid
Fahmid Al Farid
SMSifti Al Mahmud
Sifti Al Mahmud
Muntasir MomtazMuntasir Momtaz
Muntasir Momtaz
SRSHAKILA RAHMAN
SHAKILA RAHMAN
Hezerul Abdul KarimHezerul Abdul Karim
Hezerul Abdul Karim
Increased waste volume and limitations of traditional separation methods have made waste management a hot topic in recent years. To enable the recycling process to be optimized and to minimize environmental impact, waste materials must be well detected and classified. Building on this research, the system is an automated waste-detecting system that integrates machine vision and artificial intelligence (AI). It is coupled with advanced convolutional neural networks (CNNs), which are used for data collection, real-time waste detection, and classification of the proposed framework. Images of waste were captured in many different settings and analyzed with a YOLOv12-based model. The system achieves more gain in detecting and categorizing waste types with 73% precision and a mean average precision (mAP) of 78% in 100 epochs. Results indicate that the YOLOv12 model surpasses the current detection algorithms to provide an efficient and scalable solution to waste management challenges.

6 May 2024 in IEEE Access
Verified
Journal Article
Open access
SRSyed Taha Yeasin Ramadan
Syed Taha Yeasin Ramadan
TSTanjim Sakib
Tanjim Sakib
Fahmid Al FaridFahmid Al Farid
Fahmid Al Farid
MIMd Shofiqul Islam
Md Shofiqul Islam
JAJunaidi Abdullah
Junaidi Abdullah
Hezerul Abdul KarimHezerul Abdul Karim
Hezerul Abdul Karim
Global food security is seriously threatened by wheat leaf disease, which makes effective and precise disease detection and classification techniques necessary. For efficient disease control and the best possible crop health, timely identification and precise classification are essential. However, the limited availability of datasets for wheat leaf diseases hinders the development of effective and robust classification models. This research emphasizes the importance of precise wheat leaf disease diagnosis for global food security. The existing methods face challenges with limited data and computational demands. The research explores the potential of deep learning for automated disease detection, considering these challenges. CycleGAN proved to be the most effective among various augmentation techniques, enhancing the performance of classifiers DenseNet121, ResNet50V2, DenseNet169, Xception, ResNet152V2, and MobileNetV2. ADASYN also significantly improved classification accuracy, with MobileNetV2 consistently outperforming across different augmentation methods. This technique excels in overcoming challenges posed by limited datasets and class imbalances. Using CycleGAN for data augmentation notably enhanced classifier performance, addressing the scarcity of real-world samples. Evaluation through confusion matrix analysis revealed a minimal number of misclassified images—possibly as low as 0 to 3 images over the test dataset. The exceptional 100% accuracy achieved by the MobileNetV2 model on both CycleGAN and ADASYN augmented datasets highlights the potential of these techniques to unlock new levels of accuracy in wheat disease classification. This augmentation technique fine-tuned the classifier, reducing errors and highlighting the crucial role of CycleGAN in enhancing the accuracy and precision of wheat disease classification models. The proposed method establishes CycleGAN’s effectiveness in augmenting wheat leaf disease classification and recognizes ADASYN’s potential. The developed technique shows promise for automated disease detection in agriculture, enhancing global food security. Future research may optimize computational efficiency and explore integrating emerging technologies such as edge computing.

8 November 2021 in Frontiers in Pharmacology
Verified
Journal Article
Open access
AMAitana Dy Macariola
Aitana Dy Macariola
TSTheara Mae Capacion Santarin
Theara Mae Capacion Santarin
FVFerianne Joy Manday Villaflor
Ferianne Joy Manday Villaflor
LVLeofe Marie Guintos Villaluna
Leofe Marie Guintos Villaluna
RYRea Shane Leonora Yonzon
Rea Shane Leonora Yonzon
MTMyles Joshua Toledo Tan
Myles Joshua Toledo Tan
Breaking Barriers Amid the Pandemic: The Status of Telehealth in Southeast Asia and its Potential as a Mode of Healthcare Delivery in the Philippines