Zain Ali Malik

Academic

Verbatim text

Zain Ali Malik
03110099608
Phase - 8, Bahria Town, Rawalpindi
LinkedIn:
Education
Electrical Engineering
Computer Architecture, Embedded Systems, Internet of Things, Machine Learning, Operating Systems
The City School
Physics, Chemistry, Mathematics
Maria International School, Al-Jubail, KSA
Physics, Chemistry, Mathematics, Information Communication Technology
Projects
Final Project:
An Electromyography Controlled Wheelchair (ESP-32, I2C, EMG Sensor Interfacing)
Developing a muscle-controlled wheelchair to enhance mobility and independence for users. The wheelchair
includes safety features like obstacle and fall detection.
Semester Projects:
Design & Implementation of 6 stage MIPS Pipeline (MIPS Architecture, Vivado, Verilog)
Executed the integration of add, sub, lw, sw, and bne instructions using an 8-bit ALU
Digital Locker (ATMega32, UART, Sensors and Actuators integration)
Used 2 AVRs, one for serial communication and another for integrating peripherals.
Satellite Attitude Control Analysis & Design (MATLAB, PID Control)
Analyzed plant model and designed a lead controller for higher performance.
Ultrasonic Radar using Raspberry Pi Model 3B+ (Raspbian, Pygame, Ultrasonic Sensors, Motors)
Developed an ultrasonic radar with GUI display to detect object and measure distance.
Work Experience
September 2022 - May 2024
Grading assessments and assisting students in understanding complex concepts of Applied Calculus (MT-117) and
Feedback Control Systems (EE-3004)
Skills & Tools
Professional Skills:
Project Management, Problem Solving, Adaptability, Leadership, Technical Writing
Technical
Skills:
ESP-32, STM-32, Raspberry Pi, ATMega32, ESP-IDF, C/C++, Python, Verilog, Assembly, Assembly,
AWS IoT Core, Blynk, ThingSpeak, I2C, UART, SPI, Ubuntu, MATLAB, Simulink, Cisco packet Tracer,
WireShark, AutoCAD, Proteus, Git, MS Office Suite, Jira, Trello
Achievements
1st Position in HSRC'23 (Headstart School Robotics Competition) - Maze Solver Category
Mentioned in Dean's List in Fall 2023, Spring 2023 and Fall 2020
Silver Medal in Spring 2023 and Bronze Medals in Fall 2023, Fall 2020
Training / Certification
Applied AI Professional Certificate - IBM (Coursera)
Data Science Math Skills - Duke University (Coursera)
Mastering Microcontroller and Embedded Driver Development (Udemy)
Microcontroller Embedded C Programming (Udemy)
Programming with Cloud IoT Platforms - POSTEC University (Coursera)
Activities
Vicechair - IEEE/WIE, Event Cohead - NaSCon'24, Finance Secretary - ISYWSC'23
Interests
Book Reading, Movies, Travelling
i201004@nu.edu.pk
https://www.linkedin.com/in/zain-ali-malik-823369243
Majors:
ALEVELS (
)
IGCSE (
)
AI – Assisted Cardiac Patch
Patients suffering from heart problems can be diagnosed through an electrocardiogram.
However, traditional techniques mandate a hospital visit and the monitoring duration is
usually not enough to detect diseases like arrhythmia. Devices such as Holter monitors
and Zio patches are unaffordable for patients in third-world countries. Therefore, patients
need a portable device, cost-friendly, and can monitor heart activity for longer periods.
The data can be sent to the doctor to notify whether the patient is at risk of cardiac
disease.
The AI-Assisted Cardiac Patch is a device attached to the chest of a patient that
measures heart rate and breathing rate, transmits it to the cloud, stores it in a database,
and then proceeds to classify the heart disease using a machine learning model. A
webpage then visualizes the waveform and generates an alert in case of life-threatening
situations.
Technology Used:
AWS, Google collab, Arduino MATLAB,
Supervisor Name:
Dr. Mukhtar Ullah
Group Members:
Aabia Ather (i19 - 0788)
Sajjal Muddasar (i19 - 0849)
Malaika Azeem (i19 - 0886)
BAIONICS: AI-Driven EEG Brain Interface for Healthcare
BAIONICS aims to simplify the creation and implementation of Brain-Computer Interface
(BCI) applications on the edge, demonstrated with a robotic arm. BAIONICS employs the
use of non-invasive EEG signal acquisition circuitry to receive, amplify and filter required
EEG data from the brain. BAIONICS eases development of BCI applications through its
user-friendly and intuitive Graphical User Interface (GUI), in which the user can collect
their dataset and map their data to particular actuations. BAIONICS open-source nature
ensures that users can further customize their application by modifying the code as per
their own requirement.
According to the World Health Organization (WHO), 35–40 million people worldwide
require prosthetic or orthotic services BCI technology. With BAIONICS, there is an
increase in accessibility, as it lowers the barrier to entry for BCI technology. The project
also promotes enhanced innovation with it being easy to use and modify. Finally in
Healthcare applications BAIONICS provides improvement in quality of life, giving wider
access to people requiring assistive technology
Technology Used:
Python, Deep Learning, Tensorflow, Altium
Designer, SolidWorks, Raspberry Pi,
ESP32
Supervisor Name:
Dr. Niaz Ahmed
Group Members:
Adil Mubashir Chaudhry (i20 - 1001)
Usama Sadiq (i20 - 2309)
Naqi Raza (i20 - 1029)
Detection of Epilepsy Seizures in Neonatal EEG Using
Explainable AI
Epilepsy is a common neurological disorder characterized by involuntary jerking
movements and fits. These symptoms result from temporary changes in the brain's
electrical functioning. This project aims to develop an automated epileptic seizure detection
system for neonates using deep learning and explainable AI to enhance accuracy and
reliability in the diagnosis of neonatal epilepsy.
This project utilizes explainable AI because in healthcare it is addressing the need for
transparency and reliability in clinical decision making. XAI empowers healthcare
professionals to gain valuable insights into the features influencing seizure detection. This
project allows remote patient management and is beneficial for individuals with epilepsy
living in remote areas.
The project utilizes OpenBCI Ganglion board to capture EEG signals. For testing purposes,
EEG data of neonates was collected from PIMS Hospital, Islamabad. The AI model is
trained on a dataset of NICU seizures recorded at the Helsinki University Hospital. Real-
time EEG data is sent to Cloud where our AI model is deployed. The model receives raw
EEG signal, processes it and classifies it as seizure or normal activity enabling detection of
epilepsy seizure in real-time in neonatal patients. The results are displayed on a web
application.
Technology Used:
Artificial Intelligence, Machine Learning,
IoT, Python
Supervisor Name:
Dr. Muhammad Tariq
Group Members:
Omer Jamil 20I-1037
Salman Mubashir 20I-2305
Ahmed Hussain 20I-1007
FPGA Implementation of Deep Neural Network
for Surveillance Security System
This project aims to implement a trained deep convolutional neural network (CNN) on a
field-programmable gate array (FPGA) to enable fast data processing within a tiny footprint
for real-time AI applications such as smart surveillance security systems.
Deep neural networks (DNNs) are machine learning algorithms that are inspired by the
function and structure of the human brain. They encompass layers of interconnected
neurons that process and transmit the information. CNNs are one of the types of DNNs that
are utilized for computer vision tasks namely image processing plus robotic vision.
DNNs are computationally complex algorithms and thus highly power-consuming. FPGAs
are an optimal solution for implementing CNNs because of their customizability and energy
efficiency. They can be configured for any specific task, making them more power-efficient
than GPUs and ASICs. Therefore, FPGA implementation of DNNs results in power and
performance-efficient systems.
Technology Used:
MATLAB,
Vivado ML
Supervisor Name:
Dr. Farhan Khalid
Group Members:
Umm E Aiman Jalali (i19 - 0763)
Maryam Waheed (i20 - 1003)
Marryum Naeem (i20 - 1919)
Implementation of Supercomputer as a Service
The global demand for affordable, portable computers with powerful computational
capabilities is steadily increasing. As Morre's law is coming to an end and we cannot
increase the number of transistors per unit area. To address this demand, cluster
computing emerges as a viable solution. This project implements cluster computing
utilizing single-board computers, specifically the Raspberry Pi, as a cost-effective
alternative to traditional supercomputers. Our project mirrors the architecture of a
supercomputer, employing the Message Passing Interface Standard (MPI) for efficient
communication among cluster nodes.
We used MPICH, which is one of the implementations of MPI. The Message Passing
Interface (MPI) is a defined set of routines or functions that enable processes in a parallel
system to communicate with each other by sending and receiving messages. MPI creates
processes for parallel code implementation on the cluster. The master node is responsible
for dividing tasks into subparts, and after their computation, they combine their results,
while slave nodes are supposed to do the individual tasks that are being assigned.  Each
process is assigned a unique process ID or rank, helping to determine the specific task
assigned to each process. The master process gathers results from all slave nodes and
sends back the results to the end user.
This cluster will easily accessible worldwide via the web interface, allowing users to
upload their files and code for computation. The results of the computation are then
displayed on the web interface.
This device aims to complete lengthy calculations efficiently within a shorter period. The
cluster functions as a little supercomputer. The addition of more nodes enhances its
efficiency.
Technology Used:
MPICH, MongoDB, NodeJS, Raspbian
Supervisor Name:
Dr. Ata-ul-Aziz Ikram
External Supervisor Name:
Engr. Hamza Ali Imran
Group Members:
Muhammad Jawad Haleem (i20 - 2482)
Atta Ur Rehman (i20 - 1059)
Abdul Rehman (i18 - 0923)
Implementing Computer Numeric Control (CNC)
Machine to produce laser induced graphene-based Humidity
Sensor
Despite graphene's exceptional properties such as high electrical conductivity and
flexibility, current production methods are complex and costly, hindering its widespread
application.
Our project introduces a groundbreaking solution: the CNC (Computer Numeric Control)
Laser Induced Graphene Machine. By leveraging CNC technology, we've developed a
single-step, cost-effective process to produce graphene. This innovative approach
utilizes a laser to precisely etch carbon-rich materials like plastic or wood, transforming
them into graphene patterns. The CNC precision ensures accuracy and repeatability,
making this method ideal for various applications in biosensors, electro-mechanical
sensors, supercapacitors, and beyond. This advancement promises to revolutionize
graphene production, driving interest and adoption across industries.
Graphene: a single layer of carbon atoms arranged in a two-dimensional lattice,
possesses exceptional properties. In electrical engineering, graphene's exceptional
electrical conductivity is of paramount importance. Its ability to efficiently conduct
electricity allows for high-performance electronic components such as transistors and
interconnects. Additionally, graphene's flexibility enables the creation of bendable and
stretchable electronic devices, expanding the possibilities for wearable technology and
flexible displays. Furthermore, its high thermal conductivity facilitates the design of
efficient heat dissipation systems, crucial for maintaining optimal performance and
reliability in electronic devices
Technology Used:
Arduino, CNC shield and GRBL Firmware,
Laser GRBL.
Supervisor Name:
Dr. Awais Ayub
Group Members:
Asad Jamil (i20 - 2413)
Sarmad Izhar Hussain (i20 - 1017)
Maryam Naveed (i20 - 1005)
IoT-based Energy Management System Using DevOps
population is increasing. There will be a major energy crisis in the future if we do not utilize
energy efficiently. To save costs, conserve energy, and protect the environment, there is a
need to develop a system that can manage energy more effectively.
Objective:
leading to escalating electricity prices—a significant concern. Effective energy
management is imperative. This project aims to achieve this through three key
approaches:
1. Monitoring
2. Automation
3. Control
Technology Used:
Internet of things (IoT), Bluetooth Low
Energy (BLE)
Supervisor Name:
Dr. Ata-ul-Aziz
Group Members:
Mueez Ali (p20 - 0223)
Contact: 0316-1100778
Adeel Akhtar (i20 - 1025)
Contact: 0325-5461325
Obaidullah (i20-1045)
Contact: 0315-1137806
LiDAR based SLAM for creation of Digital Twins
The project aims to develop a LiDAR based SLAM system integrated with an RGB camera
to create Digital Twins. This will accurately map real-world environments and generate
visually immersive virtual replicas.
Digital Twin is one of the fourth industrial revolution advanced technologies, poised to
revolutionize product design and maintenance across various industries. The challenge
lies in creating precise Digital Twins, critical for various industries, especially in
autonomous vehicles. Inadequate Digital Twins hinder innovation and efficiency.
Our project aims to address this by developing a novel solution, enhancing accuracy and
applicability of Digital Twins, driving advancements in Lidar-based-SLAM technology, and
benefiting reliant industries. Failure to resolve this limits progress, innovation, and
competitiveness in these sectors, posing a significant obstacle.
Technology Used:
MATLAB,Python, Deep Learning,Opencv
Supervisor Name:
Dr. Muhammad Tariq
Dr. Arshad Hassan
Group Members:
Anum Zahid (i19-0759)
Abeera Abbasi (i20-1028)
Ahmad Hadi Shahzad (i20-1034)
MyoMobile : An Electromyography (EMG) Controlled
Wheelchair
The project aims to design and implement an assistive technology solution by integrating
Surface Electromyography (sEMG) for wheelchair control, empowering individuals with
limited motor functions to navigate easily and independently. The features implemented
include battery level indication and fall detection alerts for the user's safety. Furthermore,
an IoT-based mobile application allows for the remote monitoring of safety updates of the
forms of disabilities, including physical, mental, sensory, and intellectual impairments,
MyoMobile aims to empower users with greater independence and safety, bridging the gap
between conventional wheelchair technology and the specific needs of diverse user groups.
Features :
1) User Independence
2) Safety Features
3) IoT-based Mobile Application
Technology Used:
ESP-32, C, IoT ,Blynk, Human Machine
Interface
Supervisor Name:
Dr. Niaz Ahmed and Dr. Arshad Hassan
Group Members:
Neha Rizwan (i20 - 1027) –
+92 307 9266093
Samra Shahzad (i20 - 1009) - +92 324
5258544
Zain Ali Malik (i20 - 1004)-
+92 311 0099608

AI enrichment