Wade Ghribi

Wade Ghribi

Auto-generated profile, based on publicly available information

Wade Ghribi

College of Computer Science, King Khalid University, Abha, Saudi Arabia

Research Focus & Metrics

Academic Career Length

9Years (2018-Present)

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

Representative Publications

View All Publications
19 September 2024 in IEEE Access
Verified
Journal Article
Open access
DDDipshika Das
Dipshika Das
RDRudra Sankar Dhar
Rudra Sankar Dhar
PGPradip Kumar Ghosh
Pradip Kumar Ghosh
ABArindam Biswas
Arindam Biswas
SMSaurav Mallik
Saurav Mallik
WGWade Ghribi
Wade Ghribi
In this article, we present dielectric modulation-based label free detection of biomolecules at a cavity under gate metal located near the tunneling junction of charge plasma based dual metal-double gate tunnel FET (CP-DM DGTFET). The biomolecules that settle down in the cavity act as dielectric above the tunneling junction and cause drain current to increase/ decrease. To eliminate the short channel impact and to enhance I ON /I OFF ratio without compromising any other device features, dual metal gate architecture with laterally split dielectric is used. The biosensor is capable of recognizing neutral, positive, and negative charges with the highest drain current sensitivity of 2.9×10 9 Ccm -2 and highest I ON /I OFF ratio of 9.02×10 11 for biomolecule gelatin with dielectric constant k = 12. This work explores the effectiveness of charge-plasma based DM DGTFET for detection of biomolecules such as gelatin (k = 12), keratin (k = 8), bacteriophage T7 (k = 6.3), and APTES (k = 3.57) with varied charge distribution at various biasing. The drain current sensitivity, subthreshold slope, drain current, ON/OFF current ratio are computed for various biomolecules and their variations with respect to cavity length, biomolecule charge, etc. are used to detect unknown biomolecules. Finally, RF analysis such as gain bandwidth product, cut-off frequency, transit time, and linearity/selectivity of biosensors are incorporated. The device shows better linearity and less distortion for high –k biomolecules.

18 August 2025 in IEEE Access
Verified
Journal Article
Open access
DPDamodharan Palaniappan
Damodharan Palaniappan
RJRituraj Jain
Rituraj Jain
TPT. Premavathi
T. Premavathi
KPKumar Parmar
Kumar Parmar
WGWade Ghribi
Wade Ghribi
Naim AhmadNaim Ahmad
Naim Ahmad
The growing adoption of deep learning in medical imaging has propelled object detection frameworks into clinical relevance, with YOLO (You Only Look Once) standing out for its real-time inference and high accuracy. This survey offers a comprehensive review of YOLO-based medical object detection, synthesizing findings from 123 peer-reviewed papers published between 2018 and 2024. Across oncology, radiology, pathology, surgery, and PPE detection, YOLO models have demonstrated strong performance in tumor localization, lesion segmentation, tool tracking, and anomaly detection. YOLOv3 and YOLOv5 remain the most frequently deployed variants, while YOLOv8 and YOLO-NAS are emerging in recent studies with improved mAP and lower latency. Findings reveal that mAP scores exceeding 85% were commonly achieved in breast cancer and pulmonary nodule detection tasks, while lightweight versions such as YOLOv5s maintained detection speed above 50 FPS in surgical environments. Despite its clinical promise, persistent challenges include poor sensitivity to small or rare lesions, absence of benchmark datasets across modalities, and variability in annotation practices. This survey introduces a structured taxonomy of YOLO architectures, benchmarks their comparative performance, highlights key limitations, and proposes forward-looking trends involving multimodal fusion, transformer-enhanced backbones, and privacy-preserving learning. The work consolidates YOLO’s growing role in healthcare AI and serves as a foundational reference for researchers and practitioners alike.

4 March 2025 in IEEE Access
Verified
Journal Article
Open access
KCKalyan Chatterjee
Kalyan Chatterjee
MRM Raju
M Raju
Bhoomeshwar BalaBhoomeshwar Bala
Bhoomeshwar Bala
UNU. Nikitha
U. Nikitha
GPGampala Prabhas
Gampala Prabhas
SMSaurav Mallik
Saurav Mallik
Indoor air quality (IAQ) has a crucial impact on health, yet many spaces suffer from unnoticed pollution. This study introduces “Indoor Air Wellness”, a predictive model that utilizes advanced AI techniques, specifically integrating Kalman filtering (KF) and artificial neural networks (ANN), to enhance indoor air pollution prediction and management. The model excels in dynamic environments by refining the accuracy of the proposed Kal-ANN algorithm achieving a predictive accuracy (R²) of up to 96. 55%, a root mean square error (RMSE), a Mean Absolute Error (MAE), and a Mean Squared Prediction Error (MSPE) as low as 9.85, 6.12, and 3.15 g/m. ‘‘Indoor AirWellness" operates in three phases: IAQ prediction, control, and energy-efficient green building design. IAQ prediction consists of Learning and Analysis subphases, continuously monitoring and forecasting indoor parameters such as temperature, humidity, and CO levels while considering external factors. The Analysis phase implements control measures based on these predictions, leading to a 60% improvement in energy utilization in green buildings compared to traditional methods. Simulations and real-world applications demonstrate its effectiveness in reducing pollutants by up to 23% and minimizing energy consumption by up to 48%. Overall, the “Indoor Air Wellness” model enhances air quality and promotes energy efficiency, positioning it as a valuable tool for residential, commercial, and industrial applications. This study contributes to advancing smart environmental management, advocating for healthier indoor environments through AI-driven solutions.

18 November 2025 in Journal of Intelligent & Robotic Systems
Verified
Journal Article
Open access
PDF file
SSSotirios Spanogianopoulos
Sotirios Spanogianopoulos
WGWade Ghribi
Wade Ghribi
Ahmed AlkhayyatAhmed Alkhayyat
Ahmed Alkhayyat
TTTapankumar Trivedi
Tapankumar Trivedi
SRSavitha R
Savitha R
MJmostafa jalalneghad
mostafa jalalneghad
This paper investigates the precise motion control of a surgical ear robot (VTA) using a piezoelectric ultrasonic actuator (PUA). The aim is to overcome the challenges of nonlinearity, uncertainties, and disturbances in PUA. For this purpose, a non-singular terminal sliding mode controller is designed along with a disturbance observer. Simulation results show that the designed controller is able to track the desired signal with very high accuracy and in a limited time. The tracking error quickly approaches zero and no chattering is observed in the system response. The performance of the proposed controller is compared with the results of other papers. The results show that the controller proposed in this paper has a lower tracking error than the existing methods. In particular, in the steady state, the tracking error in the proposed method reaches zero, while in the comparative methods, there is an error of several micrometers. The value of the integral absolute error (IAE) for the proposed method (0.0856) is calculated to be lower than that of the methods [41] (0.0996) and [43] (1.3631), which indicates the better performance of the proposed method. Also, the investigations show that the fast terminal sliding mode controller (first design) consumes less energy for tracking than the non-singular terminal sliding mode controller (second design).

1 December 2018
Verified
Conference Paper
Closed access
AAAbdelmoty Ahmed
Abdelmoty Ahmed
RAReda Abo Alez
Reda Abo Alez
GTGamal Tharwat
Gamal Tharwat
WGWade Ghribi
Wade Ghribi
ABAhmed Said Badawy
Ahmed Said Badawy
AMAhmad M.J. Al Moustafa
Ahmad M.J. Al Moustafa
This Gestures is one of the best ways of communication between dumb and blind people depend on the expression of signs. In this paper we suggest an algorithm to recognizing hand gestures of Arabic letters to communicate between the dumb (through signs) and blind (hear the voice corresponding to sings). The proposed algorithm used the video of gesture from the dumb then convert the video into frames ( images) and calculate the distance to recognition the letters by using k-mean, k- medoid and artificial neural network, calculate the distance by using Euclidean distance and slop. There are sixteen features (8-features from Euclidean distance and 8-features from slop ). The results were (93.3673% For k-mean),(93.1354% for k-medoid ) and(92.9499% for ANN). We create our data base (from 5- videos with 308 frames).

25 March 2024 in Surface Review and Letters
Verified
Journal Article
Closed access
HYH. Asadollahi Yazdi
H. Asadollahi Yazdi
WGWade Ghribi
Wade Ghribi
MAManal A. Abbood
Manal A. Abbood
AAAmeer H. Al-Rubaye
Ameer H. Al-Rubaye
KMKarrar Hatif Mohamed
Karrar Hatif Mohamed