Muhammad Adnan

Muhammad Adnan

Auto-generated profile, based on publicly available information

Muhammad Adnan

Department of Computer Science, School of Systems and Technology, University of Management and Technology, Lahore, Pakistan

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

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

Representative Publications

View All Publications
8 January 2025 in Scientific Reports
Verified
Journal Article
Open access
PDF file
MAMohsin Akram
Mohsin Akram
MAMuhammad Adnan
Muhammad Adnan
SASyed Farooq Ali
Syed Farooq Ali
JAJAMEEL AHMAD
JAMEEL AHMAD
AYAmr Yousef
Amr Yousef
Zaffar Ahmed ShaikhZaffar Ahmed Shaikh
Zaffar Ahmed Shaikh
Deep learning-based medical image analysis has shown strong potential in disease categorization, segmentation, detection, and even prediction. However, in high-stakes and complex domains like healthcare, the opaque nature of these models makes it challenging to trust predictions, particularly in uncertain cases. This sort of uncertainty can be crucial in medical image analysis; diabetic retinopathy is an example where even slight errors without an indication of confidence can have adverse impacts. Traditional deep learning models rely on single-point predictions, limiting their ability to provide uncertainty measures essential for robust clinical decision-making. To solve this issue, Bayesian approximation approaches have evolved and are gaining market traction. In this work, we implemented a transfer learning approach, building upon the DenseNet-121 convolutional neural network to detect diabetic retinopathy, followed by Bayesian extensions to the trained model. Bayesian approximation techniques, including Monte Carlo Dropout, Mean Field Variational Inference, and Deterministic Inference, were applied to represent the posterior predictive distribution, allowing us to evaluate uncertainty in model predictions. Our experiments on a combined dataset (APTOS 2019 + DDR) with pre-processed images showed that the Bayesian-augmented DenseNet-121 outperforms state-of-the-art models in test accuracy, achieving 97.68% for the Monte Carlo Dropout model, 94.23% for Mean Field Variational Inference, and 91.44% for the Deterministic model. We also measure how certain the predictions are, using an entropy and a standard deviation metric for each approach. We also evaluated the model using both AUC and accuracy scores at multiple data retention levels. In addition to overall performance boosts, these results highlight that Bayesian deep learning does not only improve classification accuracy in the detection of diabetic retinopathy but also reveals beneficial insights about how uncertainty estimation can help build more trustworthy clinical decision-making solutions.

4 August 2022 in IEEE Access
Verified
Journal Article
Open access
KIKhalid Ijaz
Khalid Ijaz
ZHZawar Hussain
Zawar Hussain
JAJAMEEL AHMAD
JAMEEL AHMAD
SASyed Farooq Ali
Syed Farooq Ali
MAMuhammad Adnan
Muhammad Adnan
Ikramullah KhosaIkramullah Khosa
Ikramullah Khosa
An accurate electrical Short-term Load Forecasting (STLF) is an eminent factor in the power generation, electrical load dispatching and energy planning for the power supply companies, specifically in developing countries. This paper proposes a novel temporal feature selection-based Long Short-term Memory (LSTM) model developed by the combination of standard Artificial Neural Network (ANN) layer and LSTM for electrical short term load forecasting. The LSTM model has excellent capability of predicting the stochastic nature of an hour ahead electrical loads. The standard ANN layer consisting 11 neurons is used as an input to LSTM cells. Such a combination of ANN layer with LSTM was never proposed before. The proposed model accommodates variations in weather as well as temporal inputs like humidity, holidays, and date-time features in the hourly load data of the power supply company situated in Johor, Malaysia. This paper gives the insights of hyper parameter tuning to capture the more generalized electrical load patterns in the dataset without compromising the time complexity of the proposed model. The proposed approach was compared with five existing approaches, namely: ANN, LSTM model 1, LSTM model 2, LSTM model 3 and Convolutional Neural Network-LSTM (CNN-LSTM) using hourly load dataset of Johor. The experimental results demonstrate that the proposed approach outperformed the existing approaches in terms of root mean square error, mean absolute percentage error and Diebold-Mariano statistical inference test within 95% confidence interval.

27 July 2019 in Artificial Intelligence Review
Verified
Journal Article
Closed access
PDF file
SASheikh Bilal Ahmed
Sheikh Bilal Ahmed
SASyed Farooq Ali
Syed Farooq Ali
JAJAMEEL AHMAD
JAMEEL AHMAD
MAMuhammad Adnan
Muhammad Adnan
MFMuhammad Moazam Fraz
Muhammad Moazam Fraz

16 October 2023 in IEEE Access
Verified
Journal Article
Open access
GPGulnaz Perveen
Gulnaz Perveen
SASyed Farooq Ali
Syed Farooq Ali
JAJAMEEL AHMAD
JAMEEL AHMAD
SSSana Shahab
Sana Shahab
MAMuhammad Adnan
Muhammad Adnan
Ikramullah KhosaIkramullah Khosa
Ikramullah Khosa
Micro-expression recognition has gained much attention in research communities. Among its proposed solutions, deep learning approaches have shown promising results over the past few years. In this paper, we propose a multi-stream deep convolution neural network with ensemble classification for facial micro-expression recognition. The multi-stream network uses the deep features of a residual network, densely connected convolutional network, and visual geometry group. The features of these aforementioned architectures are extracted from their pooling layers and become very resource-intensive due to their high dimensions. The principal component analysis is applied to these features for their dimensionality reduction. Stacking, an ensemble classification technique, is performed on these deep features with three base learners (random tree, J48, random forest) and a meta learner (random forest). Experiments were performed using publicly available datasets, namely: CASME-II, CASME2, SMIC, and SAMM. The proposed approach (PA) is compared with twelve approaches. The results show that the PA outperformed the existing approaches in terms of accuracy and time efficiency.

1 January 2021 in IEEE Access
Verified
Journal Article
Open access
JAJAMEEL AHMAD
JAMEEL AHMAD
Muhammad ImranMuhammad Imran
Muhammad Imran
SASyed Farooq Ali
Syed Farooq Ali
MAMuhammad Adnan
Muhammad Adnan
SASyed Rehan Ashraf
Syed Rehan Ashraf
MSMuhammad Shoaib
Muhammad Shoaib
This paper provides a comprehensive assessment of wind energy potential for Hydrogen production under local conditions of Pakistan from the design, development and practical implementation perspectives. Simulations were performed for three sites—Bahawalpur, Sanghar and Gwadar using actual wind speed site data, recorded between 2016 and 2018, at intervals of 10 minute. For the selected sites, wind resource assessment was performed using the Weibull distribution function parameters, energy and wind-power density calculations at hub heights of 20m, 40m, 60m and 80m of the wind turbines. It was observed that Sanghar is the most suitable site for wind-to-Hydrogen production potential with power and energy density of $305.86~W/m^{2}$ and $2665.81~kWh/m^{2}$ , respectively. From the implementation perspective, the Nordex N90/2500 wind turbine at an 80m hub height was found to be beneficial for Sanghar with a cost of energy of 35.21 $\$ $ /MWh (0.035 $\$ $ /kWh). The cost of Hydrogen using an electrolyzer for 7-year long-term investment was 2.29 k $\$ $ /ton using Nordex N90/2500 turbine. Based on the available power density and land area, a general scheme for production of Hydrogen using electrolysis can be implemented with possibility of installation and commissioning of wind farms.

7 June 2024 in Sensors
Verified
Journal Article
Open access
PDF file
MMMuhammad Tahir Mustafa
Muhammad Tahir Mustafa
JBJung-In Baik
Jung-In Baik
SHSONG Hyoung-Kyu
SONG Hyoung-Kyu
MAMuhammad Adnan
Muhammad Adnan
WAWaqar Majeed Awan
Waqar Majeed Awan
This paper proposes a cognitive radio network (CRN)-based hybrid wideband precoding for maximizing spectral efficiency in millimeter-wave relay-assisted multi-user (MU) multiple-input multiple-output (MIMO) systems. The underlying problem is NP-hard and non-convex due to the joint optimization of hybrid processing components and the constant amplitude constraint imposed by the analog beamformer in the radio frequency (RF) domain. Furthermore, the analog beamforming solution common to all sub-carriers adds another layer of design complexity. Two hybrid beamforming architectures, i.e., mixed and fully connected ones, are taken into account to tackle this problem, considering the decode-and-forward (DF) relay node. To reduce the complexity of the original optimization problem, an attempt is made to decompose it into sub-problems. Leveraging this, each sub-problem is addressed by following a decoupled design methodology. The phase-only beamforming solution is derived to maximize the sum of spectral efficiency, while digital baseband processing components are designed to keep interference within a predefined limit. Computer simulations are conducted by changing system parameters under different accuracy levels of channel-state information (CSI), and the obtained results demonstrate the effectiveness of the proposed technique. Additionally, the mixed structure shows better energy efficiency performance compared to its counterparts and outperforms benchmarks.