Dr. Muhammad Imran Farid

CONTACT INFORMATION

Address: H#1, Lane#13, Sector E, DHA-2, Islamabad, Pakistan.      

Mobile: +92-(0) 3327788566  

Email: imran@au.edu.pk  or   drimranfarid.au@gmail.com

Status: HEC approved supervisor

CAREER SUMMARY

Date                                                                         Company                                                                                                Role

Feb 2016 - Present                                           Air University                                                                                      Assistant Professor

Sep 2010 - Aug 2011                                       Air University                                                                                      Lecturer

Apr 2009 - Jun 2010                                         Palmchip Ltd                                                                                       Senior SW Engineer (Contracted)

Sep 2008 – Mar 2009                                       COMSATS Institute of Information Technology           Research Engineer (Temporary)

Mar 2006 – Dec 2007                                       Path Intelligence Ltd                                                                      Radio Systems Assistant Engineer

Apr 2002 – Sep 2004                                         Pakistan Atomic Energy Commission                                  Scientific Officer

May 2001 – Mar 2002                                      Trust for Application of Scientific knowledge                 Research Officer

May 1999 – Mar 2001                                      Time & Tune                                                                                        Programmer 

EDUCATION

PhD       Sep 11 – Jan 16     Northwestern Polytechnical University, Xian, China. 

              (Information and Communication Engineering)  

Final thesis: Spatial and Spectral Information based Hyperspectral Image Classification. This thesis is about classification of high resolution hyperspectral images using spatial and spectral information. Hyperspectral images can be acquired from airborne or space-borne platforms, and can be used for the identification of different materials on ground as well as achieving the geometrical properties of different objects like buildings, water, roads, agriculture fields, shadows etc. This thesis focuses on the hyperspectral image (HSI) classification, which is the basis for most of the applications related to environmental monitoring and target detection.  In this thesis, some novel spatial-spectral classification approaches are proposed, which can automatically extract the useful information from hyperspectral images for different applications. Furthermore, limited availability of training samples and observed noise are common problems that many scientists encounter in practical remote sensing (RS), which are also concerned in the newly proposed classification approaches.

MSc       Oct 04 – Sep 05     University of Southampton, UK. 

              (Radio Frequency, Mobile & Sonar Communications)   (1st Div.)

Majors: MSc in Radio Frequency Communication Systems.

(Relevant subjects: Signal Processing, Sonar Systems Engineering, Digital Transmission, Cryptography and Data Compression, Mobile Radio Comm. Engineering, Radio Comm. Networks and Systems)

(Major Projects in MSc: Huffman and LZW compression and decompression in C Language. 16QAM Transceiver Design.

Final thesis: Directional finding in broadband sonar. In my thesis I simulate Narrow and Broadband Adaptive Beamforming algorithm for Sonar in MATLAB using different arrays. I did the comparison between narrow and broadband techniques and finally choose the best array for the sonar, which can detect the directional signal reception better than the other arrays in the presence of the noise. The technique can be applied to many communication systems like radar and mobile for best reception of signal quality.)

MSc        Feb 99 – Jan 01     Quaid-i-Azam University, Islamabad, Pakistan.   

              (Electronics Engineering)    (3.1)

(Relevant subjects: Applied Mathematics, Digital Logic Design, Electronics (I, II), Communication Theory, Programming Language-C, Computer Architecture, Microprocessor Based System Design, Control Systems, Software Engineering.)

BSc         Aug 96 – Jul 98     University of Punjab, Lahore, Pakistan.

          Mathematics and Physics   (3.1)

RESEARCH PAPERS

Journal Articles

[1]1. A. Khan, M. I. Siddiqui, K. I. Memon, MI Farid, et al., “Diagnostic Accuracy of Ultrasound Elastography Versus Fine-Needle Aspiration Cytology in Predicting Benign and Malignant Thyroid Nodules in Solitary and Multinodular Thyroid Glands: A Comparative Analysis,“ Cureus, vol. 16(11), 2024 (IF: 1.2)..

[2] MI Siddiqui, A Khan, KI Memon, Muhammad Imran Farid, et al., “The Role of Advanced Magnetic Resonance Imaging Sequences in Multiple Sclerosis”, Cureus, vol. 16(8), 2024 (IF: 1.2).

[3] I Khan, N Iqbal, MT Ashraf, U Farooq, Muhammad Imran Farid, et al., “Microbiomes Associated with Clinical Stages of gastric Cancer,” Biological and Clinical Sciences, vol. 5(1), 2024.

[4] M. T. Khan, Muhammad Imran Farid, et al., "Determining the Frequency of Tubal Blockage in Infertility Patients Undergoing X-Rays Hysterosalpingography," Journal of Health and Rehabilitation Research , vol. 4, no. 2, 2024. 

[5] Z. Ahmad, Muhammad Imran Farid, et al., "Magnetic Resonance Imaging Validates the Clinical Accuracy of the Mcmurray's Test in Diagnosing Medial Meniscal Tears," Pakistan Journal of Medical Health Sciences, vol. 18, no. 1, 2024. 

[6] Muhammad Imran Farid, M. T. Khalil, D. N. Ullah, et al., “Assessment of background parenchymal enhancement at dynamic contrast-enhanced MRI in predicting breast cancer recurrence risk,” Journal of Population Therapeutics and Clinical Pharmacology, vol. 31, no. 3, 2024. 

[7] S. Raza, N. Gul, H. A. Khattak, A.  Rehan, Muhammad Imran Farid, et al., “Brain tumor detection and classification using deep feature fusion and stacking concepts,” Journal of Population Therapeutics and Clinical Pharmacology, vol. 31, no. 1, 2024. 

[8] N. Sharif, N. Farid, M. Iqbal,  M. U. Haq, Muhammad Imran Farid, et al., “Gastric cancer prevention; detection of premalignant gastric lesions and surveillance for early detection of gastric cancer in Pakistan,” Journal of Population Therapeutics and Clinical Pharmacology, vol. 31, no. 3, 2024. 

[9] Muhammad Imran Farid, S. Ahmad, N. Hameed, O. F. Chidi, H. Baloch and H. U. Talha, “Effectiveness of Contrast-Enhanced Mammography in Breast Cancer Screening: A randomized control Trial”, Kurdish Studies, Vol. 12, no. 4, pp 1698-1703,  2024.

[10] A. Salam, A. Malik, N. Farid, M. Lali, Muhammad Imran Farid, et al., “Role of ultrasound in the diagnosis of non-alcoholic fatty liver disease (NAFLD),” Pakistan Journal of Medical Health Sciences, vol. 17, no. 12, 2023. 

[11] Muhammad Imran Farid, M. D. Khan, N. N. Akhtar, G. Khan, F. Ali and S. Iqbal, “Diagnosing Gastric Perforation Caused by Accidental Ingestion of Fish Bones,” Pakistan Journal of Medical Health Sciences, vol. 17, no. 09, 2023.

[12] Safdar Ali, Mehdi Hassan, Jin Yoiung Kim, Muhammad Imran Farid, Muhammad Sanaullah and Hareem Mufti, “FF-PCA-LDA: Intelligent Feature Fusion Based PCA-LDA Classification System for Plant Leaf Diseases”, Applied Sciences, MDPI, vol. 12, no. 7 : 3514, Mar. 2022. (IF: 2.838).

[13] Farid Muhammad Imran, Mingyi He and Yifan Zhang, “Systematic Comparison of Linear Feature Extraction methods for Classification of Hyperspectral Images with Noises”, International Journal of Signal Processing, Image Processing and Pattern Recognition, vol. 8, no. 9, pp. 13–20, Sept. 2015. (EI: 20154101346872). 

[14] Farid Muhammad Imran and Mingyi He, “Marker Selection using Skeletonization for very low training sample analysis of hyperspectral image classification”, International Journal of Hybrid Information Technology, vol. 8, no. 9, pp. 43–52, Sept. 2015. (EI: 20154101356229).

[15] Farid Muhammad Imran, “Marker Selection using Support Vector Machine Over-fitting for Very Low Training Sample Analysis of Hyperspectral Image Classification”, International Journal of Signal Processing, Image Processing and Pattern Recognition, vol. 8, no. 10, pp. 11–18, Oct. 2015. (EI Indexed). 

[16] Farid Muhammad Imran, “Noise Effects on Feature Mining Non-Parametric Supervised Feature Extraction Techniques”, International Journal of Multimedia and Ubiquitous Engineering, vol. 10, no. 12, pp. 207–216, Dec. 2015. (EI Indexed).

Conference Proceedings

[1] Mingyi He, Farid Muhammad Imran, Baassou Belkacem and Shaohui Mei, “Improving hyperspectral image classification accuracy using iterative SVM with spatial-spectral information”, The 1st IEEE Summit & International Conference on Signal and Information Processing (ChinaSIP), Beijing, China, pp. 471–475, Jul. 2013. (EI: 20135017076763).

[2] Farid Muhammad Imran and Mingyi He, “Extended Morphological Profiles with duality for hyperspectral Image classification”, Proceedings of IEEE WHISPERS, Tokyo, Japan, Jun. 2-5, 2015.

[3] Farid Muhammad Imran and Mingyi He, “SVM Over-fitting Marker Selection Method for Small-Sample Hyperspectral Image Classification”, Presented at Chinese WHISPERS 2015, Xi’an, pp. 126–131, Sept. 2015.

[4] Baassou Belkacem, Mingyi He, Farid Muhammad Imran and Shaohui Mei, “Hyperspectral image classification based on iterative Support vector machine by integrating spatial-spectral information”, IEEE International Geoscience and Remote Sensing Symposium (IGARSS), Australia, pp. 1023–1026, Jul. 2013, (EI: 20140917388472).

WORK EXPERIENCE

A total of seven years experience in the field of Research and Development and eleven years in the field of Teaching.

02/16 – Present   Air University, Sector E-9, PAF Complex, Islamabad, Pakistan.  

Assistant Professor 

Job responsibilities 

• • • Courses taught to Masters and Ph.D. Program:

Parallel Computing. Contents:: Introduction , Heterogeneous data parallel computing, Multidimensional grids and data, Compute architecture and scheduling, Memory architecture and data locality, Performance considerations, Convolution, Stencil, Parallel histogram, Reduction, Prefix sum (scan), Merge, Sorting, Sparse matrix computation, Graph traversal, Deep learning, Iterative magnetic resonance imaging reconstruction, Electrostatic potential map, Parallel programming and computational thinking, Programming a heterogeneous computing cluster, CUDA dynamic parallelism, Advanced practices and future evolution.

Advanced Neural Network (Deep Learning). Contents: What is deep learning? The mathematical building blocks of neural networks, Introduction to Keras and      TensorFlow, Getting started with neural networks: classification and regression, Fundamentals of machine learning and its universal workflow, Working with Keras, Deep and advanced learning for computer vision, Deep learning for Timeseries, Deep learning for text, Generative deep learning and Best practices for the real world.   

Medical Imaging. Contents: Introducing MRI The Basics, Review of Vectors, Introduction to NMR - Nuclear Magnetism, Introduction to NMR - Spin Angular Momentum and Precession,  Introduction to NMR - Longitudinal Component Magnetization,  Introduction to NMR -- Transverse Component Magnetization,  Generating an NMR Signal  Resonance and Excitation, Excitation and Frames of Reference, Introduction to Relaxation, Transverse Magnetization Relaxation, Longitudinal Magnetization Relaxation, NMR Contrast, Contrast Agents,  Introduction to Magnetic Susceptibility, The Spin Echo, Hardware – Overview, Hardware - Static Magnetic Field, Hardware - RF Coils, Hardware - Receiver (A2D), Hardware - Gradient Magnetic Fields, Slice Selection, Frequency Encoding, Phase Encoding and k-space, K-space_ Features and Tricks, Spatial Localization and k-space Review and Q&A, 3D Fourier Transform Imaging, Field of View, Receiver Bandwidth and Aliasing, Chemical Shift, Review of Spatial Localization and Bandwidth, Introduction to Pulse Sequences, The Spin Echo Pulse Sequence, Multislice Imaging,  Multiecho Spin Echo Imaging,  The Gradient Echo Pulse Sequence and Modified Flip Angle, Inversion Recovery, Saturation Techniques,  Fast Spin Echo Pulse Sequence, Fast Gradient Echo and Echoplanar Imaging, Fast Imaging Image Review, Parallel Imaging, Improving Signal to Noise ratio of an MRI, Flow effect.

Advanced Digital Image Processing. Contents: Introduction, Digital Image Fundamentals, Intensity Transformation and Spatial Filtering, Filtering in the Frequency Domain, Image Restoration and Reconstruction, Wavelets and Multiresolution Processing, Morphological Image Processing, Image Segmentation, Representation and Description (depends upon the class progress), Object Recognition (depends upon the class progress).

Satellite Imaging. Contents: Concepts and Foundations of Remote Sensing, Introduction to Visual Image Interpretation, Multispectral, Thermal, and Hyperspectral Sensing, Digital Image Interpretation and Analysis. Introduction to Astronomy.

• • • Courses taught to Bachelor of Electrical Engineering Program:

Digital Signal Processing. Contents: Processing and analysis of continuous (analog) and discrete-time (digital) signals. Sampling continuous signals: the sampling theorem, reconstruction, and aliasing. Z-Transforms and Inverse Z-Transform. Analog filters: Butterworth filters. Filter impulse and frequency responses, stability, and digital oscillators. The Discrete Fourier Transform (DFT). Fundamentals of the design and realization of the finite impulse response (FIR) and infinite impulse response (IIR) digital filters. Linear and non-linear phase filters. Decimation, interpolation, multi-rate digital signal processing.

Information Coding Theory. Contents: Importance of Information Coding Theory, Fundamentals of Information Coding Theory, Entropy and Uniquely decodable and prefix codes, Huffman Code, Kraft-McMillian Inequality, Optimal Coding, and Block Coding, Shannon Fano Elias Coding, Arithmetic Coding, Lempel-Ziv-Welch Coding, Error Control Coding Introduction, Types of Channel Codes, Decoding Strategies, Error Control Strategies,  Syndrome and error detection and correction, Encoding via convolutional code, State diagram and Trellis Diagram and Viterbi Algorithm.

Digital Logic Design. Contents:  Digital Systems and Binary Numbers, Boolean Algebra and Logic Gates,  Gate Level Minimization, Combinational logic, Synchronous Sequential Logic, Register and Counters, Memory and Programmable Logic.

Circuits and Devices. Contents:  Introduction to circuits and devices. Voltage and Current. Resistance. Ohm's Law, Power, and Energy. Series dc circuits. Parallel dc circuits. Methods of Analysis. Network Theorems. Capacitors. Inductors.

09/10 – 08/11       Air University, Sector E-9, PAF Complex, Islamabad, Pakistan.  

Lecturer 

Job responsibilities 

• • • Course taught to Bachelor of Electrical Engineering Program:

Wireless Communications. Contents:  Introduction to Wireless Communication Systems. Modern Wireless Communication Systems. The Cellular Concept--System Design Fundamentals. Mobile Radio Propagation--Large Scale Path Loss. Mobile Radio Propagation--Small Scale Path Fading and Multipath. Equalization, Diversity, and Channel Coding.  Wireless Systems and Standards--GSM.

• • • Laboratories conducted for Bachelor of Electrical Engineering Program:

Lab of Analog Electronics, Digital Logic Design, and Network Analysis. I used Proteus in Analog Electronics, Verilog in Digital Logic Design, and PSpice in Network Analysis for simulation and circuits design. 

04/09 – 06/10       Palmchip Ltd. F-10/2, Islamabad, Pakistan. (Contracted) 

Senior Software Engineer (Programming) 

Job responsibilities and projects. [Follow ISO 9001:2000]

• • • l  RFID. The project is to make the interface for RFID machine which also has fingerprint facility. My job is to make GUI in wxPython which interact with MySQL on the back end. GUI can take live employee fingerprint and RFID card data to detect a particular person. The complete data of an employee is appears on the screen including his/her picture. GUI also has the ability to add and edit new users. Other than this it can work for any number of machines means simultaneously one can see different RFID's machines information. I have also used BoxSizers so that GUI can work with all screen resolutions.

                   Apart from this I have created website of this project in Django using Apache Server interface with MySQL. The website can create and edit the users data                             and have the ability for every employee to check his/her attendance. The website has the ability to operate level wise e.g. the permissions of manager and                               branch manager is different. I have also used Java Query where required e.g. creating table. [Window XP, NetBeans 6.7]

• • • SeaRescue Project for Seagate USA. The project is to take the backup (clone) of operating system and other personal data into a hidden partition. One can update the backup anytime. The SeaRescue also restore the backup and also has the ability to set back to the original settings. It first looks for the space in OS partition, if not it then look for the available space in other partitions. It can also compress and uncompress the backup (clone).

                     In this project I have studied different hard drives and its partition formation and Program accordingly. My major work was using Linux in Python and to make                     GUI using wxPython. Using parted command I got all the information of partition I need. I use ntfsclone command to clone system and use fdisk to create and                        manage partition and use other Linux commands where required. [Fedora 10, NetBeans 6.7] 

• • • l  Created Platform Independent front end for SeaRescue.  [Fedora 10, NetBeans 6.7]

    • l  By using Revisor tool I create .ISO image of the code using the kick start file. The .ISO image is then loaded into USB using Live USB Creator or UnetBootIn Tools.

    • l  I extensively use Sun Virtual Box to simulate the project before using it on actual hard drives.

    • l  Solid-State Hard Drive Project for Sand Force USA. Converting XML into HTML via XSLT using Python.  [Fedora 10, Eclipse SDK] 

    • l  Writing different Python scripts to test solid-state hard drives, which include the ability, memory and different performance tests.  [Fedora 10, NetBeans 6.7]

    • l  Unit testing in Python for the APIs written in C and C++. [Fedora 10, NetBeans 6.7] 

    • l  Regression testing program in Python for all the scripts. [Fedora 10, NetBeans 6.7]

09/08 – 03/09      COMSATS Institute of Information Technology, Islamabad, Pakistan. (Temporary) 

Research Engineer (Research and Development). 

• • • Controlling Optical Mark Reader from PC via serial port using Python programming language and make the GUI for the client. I am using pyserial module to control serial port in Python and using wxPython module to make the GUI. The Optical Mark reader sends the data to PC using Atmel AT89C51 microcontrollor, programmed in C using Keil-IDE. [OS using Ubuntu, Eclipse SDK]  

03/06 – 12/07       Path Intelligence Ltd., Port Solent, Portsmouth, U.K.

Radio Systems Assistant Engineer (Research and Development)

Job responsibilities and projects.

• • • Tracking people via their Mobile. I implemented different algorithms like CBF, Capon, Loaded Capon, MLE, MUSIC and DOSE in Python and MATLAB to see which one performs best in outdoor environment. For indoor environment I measure the signal strength after decoding the ARFCN and TMSI from logical control channels of TDMA GSM. I have used Gnu Radio routines to see the real time signals using Universal Software Radio Peripheral. [OS used Linux SUSE, Eclipse SDK]

    • Successfully created the test database in Python of the missing information between two or more signals reception of varying time of many mobile users, based on probability methods Monte Carlos and Markov Chain.  [OS used Window XP, Eclipse SDK]

    • Program to extract the information from the database and converted it into the user defined outputs. All outputs were loaded into the .csv file and were also loaded into the Open Office Spread Sheet to make graphs using Python-UNO module. I used Python programming language on the front end and on the back end I used PostgreSQL. [SUSE, Eclipse SDK] 

    • Assembling, designing, testing and trouble-shooting the mobile directional receivers, going on field to take indoor mobile fingerprints, purchasing components and maintaining lab. 

04/02 – 09/04       Pakistan Atomic Energy Commission, Islamabad, Pakistan.

Scientific Officer (Research and Development)

Major achievement: I successfully organized a team that had to design and test in 6 months, 6 different PCB designs, including 150 tested copies of each PCB. The project was delivered on time and within budget.

 Major projects:

• • • +30V variable power supply of 5A including display panel meter.

    • +3KV variable power supply of 3mA including display panel meter. 

    • Microcontroller (AT89C52) based chronometer interface via serial port to PC.

    • Pre Amplifier, Amplifier and Discriminator for Beta and Gamma radiation detection.

05/01 – 03/02      Trust for Application of Scientific Knowledge, F-8/1, Islamabad, Pakistan.

Research Officer (Research and Development)        

Major projects:

• • • Multiple choice objective type test evaluation machine. I used Microcontroller (AT89C52) to control the machine and send data to PC via serial port. On PC I programmed in C Language to receive data.  

    • Blinking car number plate using AT89C51 Microcontroller.

05/99 – 04/01       Time & Tune, Blue Area, Islamabad, Pakistan.

Programmer (Installation and sales)

• • • Installation of different operating systems and software packages. 

    • PC trouble shooting.                                                                                                                                                                                 

THESES AND  PROJECTS SUPERVISED

 MS Theses:

• Muhammad Luqman (In progress)] Hyperspectral Brain Tumor segmentation using Graph Neural Networks.

• Asif Mehmood (In progress)] Multivariate Time-Series Earthquake Forecasting.

• Danish Zahoor (In progress): Auxiliary Learning for Multi-task Deep Neural Network.

• Tayyaba Riasat (In progress): Deep Convolutional Generative Adversarial Networks to Enhance Artificial Intelligence in Healthcare: A Skin Cancer Application.

• Urooj Fatima (Graduated):  Improving Ground Truth Data for Human Brain Tumor Diagnosis.

• Syeda Hiba Hareem Tayyab (Graduated): Efficient In-Vivo Human Brain Tumor Detection Using Wavelets. 

• Bilal Shaukat (Graduated): Improving  Binary and Multi-class Classification Accuracy of Breast Histopathology Images in Deep Learning.

• Farhan Zafar (Graduated): Fast  Algorithms for Non-Linear Unmixing with Applications in Hyperspectral Imaging (HSI).

• Adil Saeed (Graduated): Improving  Classification Performance of Hybrid EEG-FNIRS BCI System by Channel Optimization.

• Fazeel Ahmed (Graduated): Unmixing Algorithms for Application in HSI.

• Bushra Iqbal (Graduated):  Comparison of different Automatically selected markers techniques in HSI Classification.

• Maha Irshad (Graduated): Improving HSI Classification Accuracy by using SVM Overfitting incorporating unmixing techniques.

• Nasir Ali (Graduated): Morphological and Attribute Profile for Spectral and Spatial Classification of HSI.
  
  
  

Final Year projects:

• • • Breast Cancer Diagnosis using Ultrasound Imaging

    • Earth Quake Prediction Using Artificial Intelligence

    • AI Based ECG Diagnostic System

    • Detection of Lung Cancer via Deep Learning Techniques. 

    • Water Management System in Power Plants.

    • Runtime Electricity Consumption Information with SMS Feature.

    •  Designing of Charge Controller for Micro-Wind Turbine.

    •  Detection of Leukemia using microscopic images of blood samples.

    •  Smart Drawing Robot via the Internet of Things.

    •  Remotely controlled Grass Cutter Machine.

    •  Smart Attendance system via RFID.

    •  Transfer data from one PC to another wirelessly.

    •  Tracking people via their mobile in the indoor environment.                                                                                                                                                                        

SKILLS

• Languages: Strong reading, writing and speaking competencies for English and Urdu. Basic speaking level for Mandarin Chinese.

• Coding: Python, MatLab,  CUDA C, C/C++, ...

• Databases:  MySQL, PostgreSQL

• Misc: Academic research, teaching, training, consultation, LaTeX typesetting and publishing.

MISCELLANEOUS EXPERIENCE

• Awards and Achievements

Won PhD Scholarship, Faculty Development Programme,  Air University.

• Sports

Won many Prizes and Gold Medals in sports.

• Continuing professional development (CPD)

  1. 09/2024 [Graph Neural Networks (GNNs)] {Main Contents}: Introduction to Graph Neural Networks (GNNs), Understanding GNNs and its Variants, Hands-on with PyTorch Geometric and RDKit, Hands-on Node Classification on (Knowledge) Graphs, How to include edge features?, Multidimensional edge features, How edge feature are used in PyTorch Geometric.

  2. 05/2023 [Generative Adversarial Networks (GANs)]  {Main Contents}: Importance of GANs, What are GANs, Math behind GANs, A bag of tricks on GANs, Practical example on GANs, Embedding Layers, Conditional GANs (CGANs), Practical example on CGANs, Practical example om Image to Image Translation (Pix2Pix) using CGANs.

  3. 12/2022 [Data Pipelining, Multiple GPUs and Transfer Learning for Deep Learning Applications].  {Main Contents}:  What is Pipelining, Example of Pipelining in Machine Learning, Distribution Strategies for connecting multiple GPU, Example of Tensorflow hub.