Publications

Date of Award

02-2025

Document Type

Conference

Department

Computer Science

Faculty Advisor

Ahmed Khaled, Ph.D. 

Abstract

Recent advances in healthcare have provided a wide range of medical devices and biosensors that collect biomedical data from patients during different monitoring phases. At the same time, data repositories and data-oriented frameworks are needed to store, manage, analyze, and act upon the collected data to enable a wide range of healthcare services. However, storing raw medical data (e.g., hospital visits, vital signs) with no context offers incomplete information to the medical staff and facilities to efficiently offer services. Along with the raw data challenge, given the heterogeneity of the collected medical data and the medical services to be offered, utilizing one form of data repository can be challenging. In this paper, we propose layered data pipeline that targets these challenges through two main aspects. The first aspect is to build Context-aware Electronic Health Records (EHRs) that link the collected medical data with the context for a more concrete and profound image of the patient's medical status. The second aspect is to build a data pipeline that utilizes a set of databases to store, manage, and represent the EHRs and specific sub-portions to enable both patient-oriented and collective-oriented healthcare services. Patient-oriented services focus on individual EHRs for customized and tailored healthcare assistance, while collective-oriented services consider an anonymous collection of EHRs linked together to build a knowledge graph. The proposed system comprises three layers, where each layer builds parts of the context-aware EHRs and hosts certain stages of the data pipeline. The system is a work in progress, and in this paper, we discuss the main components of the proposed system, define the high-level structure of context-aware EHR, and present the proposed data pipeline through a proof-of-concept implementation for certain features.

Recommended Citation

A. Khaled and M. Bangash, "Layered Data Pipeline for Context-Aware Electronic Health Records and Healthcare Services," 2024 IEEE International Conference on E-health Networking, Application & Services (HealthCom), Nara, Japan, 2024, pp. 1-6, doi: 10.1109/HealthCom60970.2024.10880711. keywords: {Medical devices;Databases;Hospitals;NoSQL databases;Pipelines;Medical services;Knowledge graphs;Electronic medical records;Monitoring;Biomedical imaging;Electronic Health Records;Context-Aware EHR;Healthcare Services;NoSQL Databases;MongoDB;Neo4j},

Date of Award

12-2022

Document Type

Thesis

Department

Computer Science

First Advisor

Ahmed Khaled, Ph.D. 

Abstract

The medical domain is facing an ongoing challenge of how patients can share their health information and timeline with healthcare providers. This involves secure sharing, diverse data types, and formats reported by healthcare-related devices. A multilayer framework can address these challenges in the context of the Internet of Medical Things (IoMT). This framework utilizes smartphone sensors, external services, and medical devices that measure vital signs and communicate such real-time data with smartphones. The smartphone serves as an “edge device” to visualize, analyze, store, and report context- aware data to the cloud layer. Focusing on medical device connectivity, mobile security, data collection, and interoperability for frictionless data processing allows for building context-aware personal medical records (PMRs). These PMRs are then securely transmitted through a communication protocol, Message Queuing Telemetry Transport (MQTT), to be then utilized by authorized medical staff and healthcare institutions. MQTT is a lightweight, intuitive, and easy-to-use messaging protocol suitable for IoMT systems. Consequently, these PMRs are to be further processed in a cloud computing platform, Amazon Web Services (AWS). Through AWS and its services, architecting a customized data pipeline from the mobile user to the cloud allows displaying of useful analytics to healthcare stakeholders, secure storage, and SMS notifications. Our results demonstrate that this framework preserves the patient’s health-related timeline and shares this information with professionals. Through a serverless Business intelligence interactive dashboard generated from AWS QuickSight, further querying and data filtering techniques are applied to the PMRs which identify key metrics and trends.

Recommended Citation

Bangash, Muhammad, "Smartphone as an Edge for Context-Aware Real-Time Processing for Personal e-Health" (2022). University Honors Program Senior Projects. 40. https://neiudc.neiu.edu/uhp-projects/40

Date of Award

05-2022

Publication/Publisher

John S. Albazi Student Research and Creative Activities Symposium

Department

Computer Science

Faculty Sponsor 

Xiwei Wang, Ph.D. 

Abstract

In the digital era, an essential ingredient of numerous online vendors and various types of websites is a recommender system. This technology immensely reduces users’ search time when looking for the contents of their interest. Recommender systems make suggestions based on individual preferences learned from their users. Generally speaking, users have to unconditionally share personal search and/or purchase history with service providers, who also have full access to their private preferences. In this research, we have implemented a privacy-preserving point-of-interest recommender system based on a framework with three major components: a mobile app IncogniToGo, a vendor-hosted aggregate server, and a remote central server. Furthermore, we improved the system’s prediction accuracy by estimating each anonymous user’s GPS location and incorporating this information into the recommendation process. Our current model integrates the Google Cloud Platform (Maps and Firebase), a wireless communication standard called Wi-Fi Direct, and machine learning algorithms. IncogniToGo allows a user to rate a place in Google Maps. Internally, it computes the user’s current location based on their rated places and communicates this data using a random user ID via Wi-Fi Direct to the aggregator server. User groups are created on the aggregator server, and the corresponding group preferences are then sent to the Firebase (central server). Machine learning algorithms are performed on the server to extract latent features of the shared data. Finally, IncogniToGo pulls such features from Firebase and generates personalized recommendations locally on the user’s device, which prevents the server from learning users’ individual preferences.

Recommended Citation

Bangash, Muhammad, et al. “Improving the Accuracy for Privacy-Preserving Point-of-Interest Recommender Systems.” NEIU Digital Commons, 2022, neiudc.neiu.edu/srcas/2022/s15/2/. 

Date of Award

05-2022

Publication/Publisher

John S. Albazi Student Research and Creative Activities Symposium

Department

Computer Science

Faculty Sponsor

Manar Mohaisen, Ph.D. 

Abstract

The mission of the 2022 Solution Challenge is to solve for one or more of the United Nations 17 Sustainable Development Goals (SDGs) using Google technology. The United Nations aims to end poverty, ensure prosperity, and protect the planet. Ever since the COVID-19 pandemic, education and student experience has been greatly impacted in terms of low retention, low enrollment, health, and career development. Our mission is to develop a user friendly app that will seamlessly help students navigate their postsecondary education impacting good health and wellbeing, quality education. Our current scope is Northeastern Illinois University (NEIU) but can extend to other institutions. The app would serve as a tool to assist in retention, support and improve student experience and overall wellness. Our application will showcase features and user interaction will be observed with student organizations such as the GDSC NEIU chapter. The podium presentation at the symposium will include a live demo, a recorded video, and a high level overview of our mobile and cloud architectures. Furthermore, we will be continuing this project and the Q&A session will allow us to gather more insights on improving our application and its use cases.

Recommended Citation

Sanchez, Alejandra C, et al. “Improving Student Wellness and Quality Education through Google Technology.” NEIU Digital Commons, 2022, neiudc.neiu.edu/srcas/2022/s09/5/. 

Date of Award

04-2022

Publication/Publisher

In Proceedings of the 14th International Conference on Computer Supported Education

Department

Computer Science

First Author

Xiwei Wang, Ph.D. 

Abstract

As an integral component of human society, higher education has been undergoing a transformation in multiple aspects, such as administrative reorganization, pedagogical reform, and technological innovation. To line up with the latest trends, many institutions constantly update their curriculum, which poses challenges to students and their advisors. This paper proposes a machine learning-based course enrollment recommender system that aims to make personalized suggestions to students who expect to take classes in the upcoming semester.Using matrix factorization as the core algorithm, the model exploits several available types of information, including student course enrollment history and other contextual features, such as prerequisite restrictions, course meeting times, instructional methods, and course instructors. The system not only helps students but also facilitates their advisors’ work. Our experimental results show that the recommended courses were highly relevant while providing plenty of options to students.

Recommended Citation

Wang, Xiwei & Cui, Longyin & Bangash, Muhammad & Bilal, Mohammad & Rosales, Luis & Chaudhry, Wali. (2022). A Machine Learning-based Course Enrollment Recommender System. 436-443. 10.5220/0011109100003182. 

Date of Award

07-2021

Publication/Publisher

NEIU Student Research and Creative Activities Symposium

Department

Computer Science

Faculty Sponsor

Ahmed Khaled, Ph.D. 

Abstract

In this modern era, we have smartphones, smart homes, and smart appliances. This comes under the different scales of the Internet of Things (IoT), from the small personal smart space scale to the huge smart city scale. They are many communication protocols used by the wide variety of smart devices and applications in order to exchange data and commands in these smart spaces. Out of the many IoT protocols, there is one that is called Message Queuing Telemetry Transport (MQTT). MQTT is lightweight, intuitive, and easy to use messaging protocol suitable for IoT applications and devices. Aside from the traditional request/response communication, MQTT protocol uses a paradigm known as publish/subscribe. Users (applications/devices) can publish or send data under certain topics. Similarly, other users can show interest in receiving updates and messages published on certain topic(s) by first subscribing. A topic is a meta-data and a short description of the communicated messages (e.g., Chicago weather, traffic updates), and users can subscribe to a wide range of topics. This semester, I am adapting a 300-level course to an honors credit, fulfilling one of the curriculum requirements of the Honors Program. To adapt a course, I am expanding on what I am learning for my Computer Networks class by exploring the work of the MQTT protocol to implement a user-friendly JavaScript-based dashboard for MQTT users. Using the dashboard, a user can subscribe to a topic or a set of topics, publish a message(s), subscribe to topic(s), and view the published messages in a suitable format. The developed dashboard allows different types of messages to be published and displayed, that include JSON/XML-based content, documents, and images. The presentation in the symposium will include a live demo running multiple instances of the dashboard, representing the different users, for the attendees to participate and test the real-time message exchange through the MQTT dashboard. 

Recommended Citation

Bangash, Muhammad, and Ahmed Khaled. “Implementing a MQTT Client Dashboard -- an IoT Protocol.” NEIU Student Research and Creative Activities Symposium, 22 Apr. 2021, neiudc.neiu.edu/srcas/2021/s33/1/.