MGMona Fazel Ghaziani

Mona Fazel Ghaziani

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Mona Fazel Ghaziani has a total of 31 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    A lightweight deep learning pipeline for MRI-based differential diagnosis of Alzheimer’s disease and mild cognitive impairment

    by Zahra Hasankhani, Gelareh Valizadeh, Davoud Khezerloo, Hamidreza Saligheh Rad, Mona Fazel Ghaziyani

    Background: Alzheimer’s disease (AD) accounts for 60–70% of dementia cases, with mild cognitive impairment (MCI) as a prodromal stage. Early diagnosis is critical. This study aimed to develop and validate a lightweight deep learning pipeline for automated classification of AD, MCI, and cognitively normal (CN) using routine 1.5 T MRI. Method: We analyzed 567 ADNI participants (95 CN, 368 MCI, 104 AD ). Preprocessing of T1-weighted MRI included skull stripping, resampling, and intensity normalization. Eleven consecutive coronal slices covering the hippocampus were extracted per subject. Two CNNs (ConvNeXt-Tiny, EfficientNet-B0) were trained on a stratified 80/10/10 patient-level split. Binary classification tasks: AD vs CN and AD vs MCI. Result: ConvNeXt-Tiny achieved AUCs of 0.98 (AD vs. CN) and 0.74 (AD vs. MCI), outperforming EfficientNet-B0 (0.98 and 0.66, respectively). Sensitivity/specificity for AD vs CN were 0.94/0.96; for AD vs MCI, 0.68/0.71. Conclusion: CNN-based classification of 1.5 T MRI was feasible for AD vs CN, but performance dropped substantially in AD vs MCI, which suggested a need for more discriminative features or multimodal input. The lightweight pipeline was suitable for integration into routine neuroimaging practice.

  • Open Access

    Prediction of the occurrence of death in East Azerbaijan province road accidents using the traffic monitoring cameras data: An application of artificial intelligence

    by Reza Arabi Belaghi, Neda Gilani, Homayoun Sadeghi-Bazargani, Aysan Mohammad-Namdar, Nasim Hajipoor Kashgsaray, Alireza Razzaghi, Mona Fazel Ghaziani

    Introduction: Road traffic injuries (RTIs) are one of the most important public health problems and causes of mortality worldwide, and especially in Iran. Methods: We used data from 2017-03-19 to 2021-03-20 registered in RTIs by the East Azerbaijan forensic medicine organization database. Information on predictor variables was obtained from traffic monitoring cameras’ data. We developed eight machine learning prediction models: logistic regression (LR), elastic net regression, decision tree (DT), random forest (RF), extreme gradient boosting (EGB), support vector machines (SVM; linear and non-linear), and artificial neural networks (ANNs). We used RF to evaluate the importance of each predictor in the prediction of death. Results: The mean number of classes 1, 2, and 4 vehicles on the road on days when death occurred was significantly higher than on days without death and there was an opposite significant pattern for vehicle types 3 and 5. Similar to the training data, RF provided the highest prediction accuracy with an AUC of 91% (95% CI:88%-93%) in the testing data. The total number of type 2 vehicles on the roads is by far the most important and relevant predictor variable (variable importance:83.95) followed by the number of instances of unsafe distance while driving (58.50). The number of Class 4 vehicles (56.58%) and average speed of vehicles (56.31%) were the next most important variables. Conclusion: Using the RF machine learning algorithm, the occurrence of death in accidents can be predicted with very high accuracy using the number of class 2 vehicles on roads.

  • Open Access

    Longitudinal assessment of MRI-based radiomics feature changes in caudate and putamen over progression of parkinson’s disease

    by Nesa Seify, Davood Khezerloo, Amir Reza Jahanshahi, Mona Fazel Ghaziani, Melika Rahimiyan

    Background: Parkinson’s disease (PD) is the second most common neurological disorder, characterized by both motor and non-motor symptoms. Recent studies have shown that the nigrostriatal pathway undergoes changes as Parkinson's progresses. Radiomics, a quantitative image analysis technique, has recently gained attention in clinical issue. This longitudinal study aimed to evaluate radiomics features extracted from the bilateral caudate and putamen during Parkinson’s progression. Methods: Twenty PD and eighteen healthy control (HC) subjects with two-time-point MRI scans selected from Parkinson’s disease Progression Markers Initiative (PPMI) database, thirty-two features were extracted between the Parkinson's and healthy groups over two-year follow-up. Results: Among these thirty-two features, eighteen significant radiomics features (P < 0.05) between PD and HC groups showed meaningful alteration over two-year follow-up. The left caudate exhibited 12 significant features, surpassing the left/right putamen and right caudate, and its significance levels were also considerably higher [p-value = 0.001- 0.008]. Conclusion: Several radiomics features in the caudate and putamen change significantly during Parkinson's progression, specifically in the left caudate. Our features that based on a quantification of the pattern of gray level or gray-level runs of voxels, reflecting gray scale heterogeneity/uniformity, offer diagnostic information beyond naked eyes detection and may reflect pathophysiology. Radiomics textural changes in left caudate may reflect pathophysiology, serving as a noninvasive biomarker for disease management.


  • Open Access

    Image quality and radiation dose in coronary CT: a systematic review and meta-analysis of deep learning reconstruction

    by Alireza Soleymanzadeh, Fatemeh Salehnia, Davod Khezerlo, Mona Fazel Ghaziyani

    Background: Coronary artery disease (CAD) is a leading global cause of mortality. Coronary computed tomography angiography (CCTA) is a widely used non-invasive imaging modality for diagnosing CAD and assessing atherosclerotic plaques. Deep learning-based image reconstruction (DLR) has emerged as an advanced technique with the potential to enhance image quality and reduce radiation dose compared to conventional reconstruction methods. Main text: This systematic review aimed to evaluate DLR’s effects on image quality and radiation dose in CCTA, comparing it to iterative reconstruction (IR) and filtered back projection (FBP). We conducted a systematic literature search across PubMed, Web of Science, Science Direct, and Embase databases, including articles up to February 2024. After screening 281 initially identified studies, 11 met our predefined inclusion criteria and underwent critical appraisal using the JBI Critical Appraisal Checklist for Diagnostic Test Accuracy Studies. We extracted and analyzed data on image noise, signal-to-noise ratio (SNR), and contrast-to-noise ratio (CNR). Our findings indicate that DLR significantly improves image quality by reducing noise and increasing both SNR and CNR when compared to IR and FBP. Moreover, DLR enables radiation dose reduction while maintaining diagnostic image quality. Conclusion: Deep learning-based image reconstruction demonstrates substantial potential for simultaneously improving image quality and reducing radiation dose in CCTA. However, further comparative studies are crucial to fully evaluate its impact on diagnostic accuracy and optimize its clinical implementation.

  • Radiomics reproducibility in computed tomography through changes of ROI size, resolution, and hounsfield unit: A phantom study

    by Y. Soleymani, Z. Valibeiglou, M. Fazel Ghaziani, A. Jahanshahi, D. Khezerloo


  • Open Access

    Radiomics reproducibility challenge in computed tomography imaging as a nuisance to clinical generalization: a mini-review

    by Amirreza Jahanshahi, Yunus Soleymani, Mona Fazel Ghaziani, Davood Khezerloo

    Background: Radiomics has demonstrated striking potential in accurate cancer diagnosis but still needs strengthening of validity and standardization to achieve reproducible and generalizable results. Despite the advantages of radiomics, inter-scanner and intra-scanner variations of computed tomography (CT) scanning parameters can affect the reproducibility of its results. Accordingly, this article aims to review the impact of CT scanning parameters on the reproducibility of radiomics results. Main body of the In general, radiomics results are sensitive to changes in the noise level; therefore, any parameter that affects image noise, such as kilovoltage (kVp), tube current (mAs), slice thickness, spatial resolution, image reconstruction algorithm, etc., can affect radiomics results. Also, region of interest (ROI) segmentation is another fundamental challenge in reducing radiomics reproducibility. Studies showed that almost all scanning parameters affect the reproducibility of radiomics. However, some robust features are reproducible. Short conclusion: One of the solutions to overcome the radiomics reproducibility challenge is the standardization of imaging protocols according to noise level (not scanning protocols). The second solution is to list reproducible features according to the type of complication and anatomical region. Resampling may also overcome feature instability.


  • Open Access

    Evaluation of textural-based radiomics features for differentiation of COVID-19 pneumonia from non-COVID pneumonia

    by Yunus Soleymani, Amir Reza Jahanshahi, Maryam Hefzi, Mona Fazel Ghaziani, Amin Pourfarshid, Davood Khezerloo

    Background: The false-positive rate of computed tomography (CT) images in the diagnosis of coronavirus disease 2019 (COVID-19) is a challenge for the management in the pandemic. The main purpose of this study is to investigate the textural radiomics features on chest CT images of COVID-19 pneumonia patients and compare them with those of non-COVID pneumonia. This is a retrospective study. Some textural radiomics features were extracted from the CT images of 66 patients with COVID-19 pneumonia and 40 with non-COVID pneumonia. For radiomics analysis, the regions of interest (ROIs) were manually identified inside the pulmonary ground-glass opacities. For each ROI, 12 textural features were obtained and, then, statistical analysis was performed to assess the differences in these features between the two study groups. Results: 8 of the 12 texture features demonstrated a significant difference (P < 0.05) in two groups, with COVID-19 pneumonia lesions tending to be more heterogeneous in comparison with the non-COVID cases. Among the 8 significant features, only two (homogeneity and energy) were found to be higher in non-COVID cases. Conclusions: Textural radiomics features can be used for differentiating COVID-19 pneumonia from non-COVID pneumonia, as a non-invasive method, and help with better prognosis and diagnosis of COVID-19 patients.


  • Open Access

    Image Quality and Dose Comparison of Single-Energy CT (SECT) and Dual-Energy CT (DECT)

    by Ramin Ghasemi Shayan, Maryam Oladghaffari, Fakhrosadat Sajjadian, Mona Fazel Ghaziyani

    CT and its comprehensive usage have become one of the most indispensable components in medical field especially in the diagnosis of several diseases. SECT and DECT have developed CT diagnostic potentials in several means. In this review article we have discussed the basic principles of single-energy and dual-energy computed tomography and their important physical differences which can cause better diagnostic evaluation. Moreover, different organs diagnostic evaluations through single-energy and dual-energy computed tomography have been discussed. Conventional or single-energy CT (SECT) uses a single polychromatic X-ray beam (ranging from 70 to 140 kVp with a standard of 120 kVp) emitted from a single source and received by a single detector. The concept of dual-energy computed tomography (DECT) is almost as old as the CT technology itself; DECT initially required substantially higher radiation doses (nearly two times higher than those employed in single-energy CT) and presented problems associated with spatial misregistration of the two different kV image datasets between the two separate acquisitions. The basic principles of single-energy and dual-energy computed tomography and their important physical differences can cause better diagnostic evaluation. Moreover, different organs diagnostic evaluations through single-energy and dual-energy computed tomography have been discussed. According to diverse data and statistics it is controversial to definitely indicate the accurate comparison of image quality and dose amount.


  • Open Access

    Anti-CD24 bio Modified PEGylated Gold Nanoparticles as Targeted Computed Tomography Contrast Agent

    by Mona Fazel Ghaziyani, Mohammd Pourhassan Moghaddam, Daryoush Shahbazi-Gahrouei, Mostafa Ghavami, Ali Mohammadi, Mehran Mesgari Abbasi, Behzad Baradaran

    Purpose: Molecular imaging is one of the import methods for recognition of cancer at the early stage in order to enhance the capacity of remedy. This study was aimed to introduce a new contrast agent that was targeted with CD24 so as to improve the CT scan detection of cancer cells with higher CD24 expression. Methods: The surface modifications of gold nanoparticles (Au-NPs) were done with long PEG (HS-PEG-CH3O) and short PEG (HS-PEG-COOH) chains to enhance their stability and capacity for immobilization of different antibodies. MTT assay was carried out to assess the biocompatibility of the NPs. The obtained contrast agent was implemented in the targeted CT imaging based on in vitro and in vivo studies of breast cancer. Results: The results revealed that the attached CD24 to the cell surface of PEGylated Au-NPs could enhance significantly the cells CT number (40.45 HU in 4T1, while it was 16.61 HU in CT26) It was shown that the attenuation coefficient of the molecularly targeted cells was more than 2 times excessive than the control groups. Further, the tumor region in model of xenograft tumor has higher density compare to the omnipaque groups, 60 min after injection (45 Hu vs.81 Hu). These results showed that the nanoparticles stayed in tumor region for longer time. Conclusion: It is predicted that the synthesized nanoparticle can be used as computed tomography contrast agent. Also, it can be used to identify the tumor cells with higher expression of CD24 at the early stages more efficiently compare to the other routine methods.