Zahra Hafezi Kafshgari

Dr. Zahra Hafezi Kafshgari


Research Focus & Metrics

Academic Career Length

10Years (2017-Present)

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

Representative Publications

View All Publications

1 November 2017
Verified
Conference Paper
Closed access
VHVahid Hassanpour
Vahid Hassanpour
SRSedighe Rajabi
Sedighe Rajabi
ZSZeinab Shayan
Zeinab Shayan
ZKZahra Hafezi Kafshgari
Zahra Hafezi Kafshgari
MAMohammad Mehdi Arefi
Mohammad Mehdi Arefi
Smart home creates the environment that maximizes the quality of life beside the efficient use of energy resources and provides the energy management systems (EMS). Since the energy crisis is a global problem, home automation should be pervasive all around the world as an important component in reducing energy and using renewable energy. One of the fundamental issues in home automation is the cost of automation. Thereby reducing the cost of automation is an important concern in the world. In this paper, a low-cost home automation system based on Arduino microcontroller has been introduced that works with Modbus protocol.

1 December 2023
Verified
Book Chapter
Closed access
PDF file
CKChamani Shiranthika Jayakody Kankanamalage
Chamani Shiranthika Jayakody Kankanamalage
ZKZahra Hafezi Kafshgari
Zahra Hafezi Kafshgari
PSParvaneh Saeedi
Parvaneh Saeedi
IBIvan V. Bajić
Ivan V. Bajić

15 January 2026 in Bioengineering
Verified
Journal Article
Open access
PDF file
CKChamani Shiranthika Jayakody Kankanamalage
Chamani Shiranthika Jayakody Kankanamalage
ZKZahra Hafezi Kafshgari
Zahra Hafezi Kafshgari
HHHadi Hadizadeh
Hadi Hadizadeh
PSParvaneh Saeedi
Parvaneh Saeedi
Machine Learning (ML) and Deep Learning (DL) have shown significant promise in healthcare, particularly in medical image segmentation, which is crucial for accurate disease diagnosis and treatment planning. Despite their potential, challenges such as data privacy concerns, limited annotated data, and inadequate training data persist. Decentralized learning approaches such as federated learning (FL), split learning (SL), and split federated learning (SplitFed/SFL) address these issues effectively. This paper introduces “MedSegNet10,” a publicly accessible repository designed for medical image segmentation using split-federated learning. MedSegNet10 provides a collection of pre-trained neural network architectures optimized for various medical image types, including microscopic images of human blastocysts, dermatoscopic images of skin lesions, and endoscopic images of lesions, polyps, and ulcers. MedSegNet10 implements SplitFed versions of ten established segmentation architectures, enabling collaborative training without centralizing raw data and labels, reducing the computational load required at client sites. This repository supports researchers, practitioners, trainees, and data scientists, aiming to advance medical image segmentation while maintaining patient data privacy.

2 July 2020 in medRxiv
Verified
Preprint
Open access
PDF file
ZKZahra Hafezi Kafshgari
Zahra Hafezi Kafshgari
MSMohammad Sabouri
Mohammad Sabouri
MSMilad Shayan
Milad Shayan
SPShahram Paydar
Shahram Paydar
Introduction: Knowing the final status of trauma patients and clearly understanding their condition are always significant due to the complexity of injuries and the high dependence of the patient’s condition on various factors. That is why it allows doctors to be able to provide required facilities in a broader perspective and perform appropriate action. In addition, it can avoid wasting time and energy and then increasing patient mortality. While, there are several ways to measure and predict the final status of patients but all of them have some defects. Therefore, it is very important for the significance of the system design with high accuracy and reliability to be able to help physicians investigate the final status of trauma patients. Method: In this study, a method, a sub-branch of data science and artificial intelligence, is presented and studied based on artificial neural networks to estimate the final status of trauma patients and predict their survival and death probabilities during the treatment and care process. In the proposed method, the final status of patients is predicted using 13 respiratory indices. This method is run in Matlab and its efficiency is studied. Results: Research subjects include 3073 patients, 494 females and 2579 males, from Shahid Rajaei Medical Center in Shiraz. In general, according to the results from testing the method, it has been able to accurately predict the mortality of patients based on respiratory indices. The proposed structure has been able to predict patients’ survival and death probabilities with an accuracy of %73.75 and %99.71 respectively. Therefore, we can conclude that the presented and examined method can make a significant relevance between calculated respiratory indices and final status of patients. Discussion and Conclusion: Due to the present study and the obtained results and investigating the mortality relevance with the other 13 respiratory indices using artificial intelligence-based methods, it can be stated that these indicators are good criteria for predicting mortality.