Pooya Iranpour
@sums.ac.ir
Radiology Department
Shiraz University of Medical Sciences
Scopus Publications
Scholar Citations
Scholar h-index
Scholar i10-index
Scopus Publications
Mahvash Dehghankhold, Fatemeh Ahmadi, Navid Nezafat, Mehdi Abedi, Pooya Iranpour, Amirreza Dehghanian, Omid Koohi-Hosseinabadi, Amin Reza Akbarizadeh, and Zahra Sobhani
Elsevier BV
Isaac Shiri, Yazdan Salimi, Abdollah Saberi, Masoumeh Pakbin, Ghasem Hajianfar, Atlas Haddadi Avval, Amirhossein Sanaat, Azadeh Akhavanallaf, Shayan Mostafaei, Zahra Mansouri,et al.
Wiley
AbstractTo derive and validate an effective machine learning and radiomics‐based model to differentiate COVID‐19 pneumonia from other lung diseases using a large multi‐centric dataset. In this retrospective study, we collected 19 private and five public datasets of chest CT images, accumulating to 26 307 images (15 148 COVID‐19; 9657 other lung diseases including non‐COVID‐19 pneumonia, lung cancer, pulmonary embolism; 1502 normal cases). We tested 96 machine learning‐based models by cross‐combining four feature selectors (FSs) and eight dimensionality reduction techniques with eight classifiers. We trained and evaluated our models using three different strategies: #1, the whole dataset (15 148 COVID‐19 and 11 159 other); #2, a new dataset after excluding healthy individuals and COVID‐19 patients who did not have RT‐PCR results (12 419 COVID‐19 and 8278 other); and #3 only non‐COVID‐19 pneumonia patients and a random sample of COVID‐19 patients (3000 COVID‐19 and 2582 others) to provide balanced classes. The best models were chosen by one‐standard‐deviation rule in 10‐fold cross‐validation and evaluated on the hold out test sets for reporting. In strategy#1, Relief FS combined with random forest (RF) classifier resulted in the highest performance (accuracy = 0.96, AUC = 0.99, sensitivity = 0.98, specificity = 0.94, PPV = 0.96, and NPV = 0.96). In strategy#2, Recursive Feature Elimination (RFE) FS and RF classifier combination resulted in the highest performance (accuracy = 0.97, AUC = 0.99, sensitivity = 0.98, specificity = 0.95, PPV = 0.96, NPV = 0.98). Finally, in strategy #3, the ANOVA FS and RF classifier combination resulted in the highest performance (accuracy = 0.94, AUC =0.98, sensitivity = 0.96, specificity = 0.93, PPV = 0.93, NPV = 0.96). Lung radiomic features combined with machine learning algorithms can enable the effective diagnosis of COVID‐19 pneumonia in CT images without the use of additional tests.
Isaac Shiri, Yazdan Salimi, Nasim Sirjani, Behrooz Razeghi, Sara Bagherieh, Masoumeh Pakbin, Zahra Mansouri, Ghasem Hajianfar, Atlas Haddadi Avval, Dariush Askari,et al.
Wiley
AbstractBackgroundNotwithstanding the encouraging results of previous studies reporting on the efficiency of deep learning (DL) in COVID‐19 prognostication, clinical adoption of the developed methodology still needs to be improved. To overcome this limitation, we set out to predict the prognosis of a large multi‐institutional cohort of patients with COVID‐19 using a DL‐based model.PurposeThis study aimed to evaluate the performance of deep privacy‐preserving federated learning (DPFL) in predicting COVID‐19 outcomes using chest CT images.MethodsAfter applying inclusion and exclusion criteria, 3055 patients from 19 centers, including 1599 alive and 1456 deceased, were enrolled in this study. Data from all centers were split (randomly with stratification respective to each center and class) into a training/validation set (70%/10%) and a hold‐out test set (20%). For the DL model, feature extraction was performed on 2D slices, and averaging was performed at the final layer to construct a 3D model for each scan. The DensNet model was used for feature extraction. The model was developed using centralized and FL approaches. For FL, we employed DPFL approaches. Membership inference attack was also evaluated in the FL strategy. For model evaluation, different metrics were reported in the hold‐out test sets. In addition, models trained in two scenarios, centralized and FL, were compared using the DeLong test for statistical differences.ResultsThe centralized model achieved an accuracy of 0.76, while the DPFL model had an accuracy of 0.75. Both the centralized and DPFL models achieved a specificity of 0.77. The centralized model achieved a sensitivity of 0.74, while the DPFL model had a sensitivity of 0.73. A mean AUC of 0.82 and 0.81 with 95% confidence intervals of (95% CI: 0.79–0.85) and (95% CI: 0.77–0.84) were achieved by the centralized model and the DPFL model, respectively. The DeLong test did not prove statistically significant differences between the two models (p‐value = 0.98). The AUC values for the inference attacks fluctuate between 0.49 and 0.51, with an average of 0.50 ± 0.003 and 95% CI for the mean AUC of 0.500 to 0.501.ConclusionThe performance of the proposed model was comparable to centralized models while operating on large and heterogeneous multi‐institutional datasets. In addition, the model was resistant to inference attacks, ensuring the privacy of shared data during the training process.
Fatemeh Farjadian, Zahra Faghih, Maryam Fakhimi, Pooya Iranpour, Soliman Mohammadi-Samani, and Mohammad Doroudian
MDPI AG
This study presents the synthesis of glucosamine-modified mesoporous silica-coated magnetic nanoparticles (MNPs) as a therapeutic platform for the delivery of an anticancer drug, methotrexate (MTX). The MNPs were coated with mesoporous silica in a templated sol–gel process to form MNP@MSN, and then chloropropyl groups were added to the structure in a post-modification reaction. Glucosamine was then reacted with the chloro-modified structure, and methotrexate was conjugated to the hydroxyl group of the glucose. The prepared structure was characterized using techniques such as Fourier transform infrared (FT-IR) spectroscopy, elemental analysis (CHN), field emission scanning electron microscopy (FESEM), transmission electron microscopy (TEM), dynamic light scattering (DLS), a vibrating sample magnetometer (VSM), and X-ray diffraction (XRD). Good formation of nano-sized MNPs and MNP@MSN was observed via particle size monitoring. The modified glucosamine structure showed a controlled release profile of methotrexate in simulated tumor fluid. In vitro evaluation using the 4T1 breast cancer cell line showed the cytotoxicity, apoptosis, and cell cycle effects of methotrexate. The MTT assay showed comparable toxicity between MTX-loaded nanoparticles and free MTX. The structure could act as a glucose transporter-targeting agent and showed increased uptake in cancer cells. An in vivo breast cancer model was established in BALB/C mice, and the distribution of MTX-conjugated MNP@MSN particles was visualized using MRI. The MTX-conjugated particles showed significant anti-tumor potential together with MRI contrast enhancement.
Devaki Shilpa Surasi, Matthias Eiber, Tobias Maurer, Mark A. Preston, Brian T. Helfand, David Josephson, Ashutosh K. Tewari, Diederik M. Somford, Soroush Rais-Bahrami, Bridget F. Koontz,et al.
Elsevier BV
Nastaran Khalili, P. Iranpour, Neda Khalili and S. Haseli
Hydatid disease is a zoonotic infection caused primarily by the tapeworm parasite, Echinococcus granulosus. It is considered an endemic disease in the Mediterranean region. In about 90% of cases, hydatid cysts are found in the liver and lungs; however, any other organ in the body may be affected, particularly in endemic areas. When encountering cystic lesions in these areas, the physician should always keep hydatid disease as a possible diagnosis in mind. To avoid life-threatening conditions such as anaphylactic shock or pressure effect on vital organs, timely diagnosis, and proper management are critical. When a rare site is involved, hydatid disease should be diagnosed using a combination of serologic assays and imaging modalities such as ultrasonography, computed tomography (CT), and magnetic resonance imaging (MRI). These imaging modalities can also be used to determine the extent of the disease and assess possible complications. Here, we present a pictorial review of typical imaging manifestations of hydatid cysts in unusual sites. Being aware of these imaging features will assist physicians in making an accurate, timely diagnosis and subsequently, providing optimal management.
Khatereh Zarkesh, Reza Heidari, Pooya Iranpour, Negar Azarpira, Fatemeh Ahmadi, Soliman Mohammadi-Samani, and Fatemeh Farjadian
Elsevier BV
Mehdi Ghaderian Jahromi, Abdolrahim Sadeghi Yakhdani, Mahdi Saeedi-Moghadam, and Pooya Iranpour
Elsevier BV
Fariba Zarei, Banafsheh Zeinali-Rafsanjani, Pooya Iranpour, and Sepideh Sefidbakht
Elsevier BV
Fatemeh Yarmahmoodi and Pooya Iranpoor
Elsevier BV
Mehdi Ghaderian Jahromi, Sara Haseli, Pooya Iranpour, and Amir Mohammad Nourizadeh
Elsevier BV
Khanh Bao Tran, Justin J Lang, Kelly Compton, Rixing Xu, Alistair R Acheson, Hannah Jacqueline Henrikson, Jonathan M Kocarnik, Louise Penberthy, Amirali Aali, Qamar Abbas,et al.
Elsevier BV
F. Zarei, R. Jalli, S. Chatterjee, R. Ravanfar Haghighi, P. Iranpour, Vani Vardhan Chatterjee and Sedigheh Emadi
Background: The present study aimed to evaluate the effectiveness of ultra-low-dose (ULD) chest computed tomography (CT) in comparison with the routine dose (RD) CT images in detecting lung lesions related to COVID-19. Methods: A prospective study was conducted during April-September 2020 at Shahid Faghihi Hospital affiliated with Shiraz University of Medical Sciences, Shiraz, Iran. In total, 273 volunteers with suspected COVID-19 participated in the study and successively underwent RD-CT and ULD-CT chest scans. Two expert radiologists qualitatively evaluated the images. Dose assessment was performed by determining volume CT dose index, dose length product, and size-specific dose estimate. Data analysis was performed using a ranking test and kappa coefficient (κ). P<0.05 was considered statistically significant. Results: Lung lesions could be detected with both RD-CT and ULD-CT images in patients with suspected or confirmed COVID-19 (κ=1.0, P=0.016). The estimated effective dose for the RD-CT protocol was 22-fold higher than in the ULD-CT protocol. In the case of the ULD-CT protocol, sensitivity, specificity, accuracy, and positive predictive value for the detection of consolidation were 60%, 83%, 80%, and 20%, respectively. Comparably, in the case of RD-CT, these percentages for the detection of ground-glass opacity (GGO) were 62%, 66%, 66%, and 18%, respectively. Assuming the result of real-time polymerase chain reaction as true-positive, analysis of the receiver-operating characteristic curve for GGO detected using the ULD-CT protocol showed a maximum area under the curve of 0.78. Conclusion: ULD-CT, with 94% dose reduction, can be an alternative to RD-CT to detect lung lesions for COVID-19 diagnosis and follow-up. An earlier preliminary report of a similar work with a lower sample size was submitted to the arXive as a preprint. The preprint is cited as: https://arxiv.org/abs/2005.03347
Isaac Shiri, Yazdan Salimi, Masoumeh Pakbin, Ghasem Hajianfar, Atlas Haddadi Avval, Amirhossein Sanaat, Shayan Mostafaei, Azadeh Akhavanallaf, Abdollah Saberi, Zahra Mansouri,et al.
Elsevier BV
Fereshteh Yazdanpanah, Pooya Iranpour, Sara Haseli, Maryam Poursadeghfard, and Fatemeh Yarmahmoodi
Elsevier BV
Hossein Abdolrahimzadeh Fard, Salahaddin Mahmudi-Azer, Qusay Abdulzahraa Yaqoob, Golnar Sabetian, Pooya Iranpour, Zahra Shayan, Shahram Bolandparvaz, Hamid Reza Abbasi, Shiva Aminnia, Maryam Salimi,et al.
Elsevier BV
Farshad Farzadfar, Mohsen Naghavi, Sadaf G Sepanlou, Sahar Saeedi Moghaddam, William James Dangel, Nicole Davis Weaver, Arya Aminorroaya, Sina Azadnajafabad, Sogol Koolaji, Esmaeil Mohammadi,et al.
Elsevier BV
R. Jalli, F. Zarei, S. Chatterjee, R. R. Haghighi, Alireza Novshadi, P. Iranpour, S. Sefidbakht and V. Chatterjee
Background: The present study was conducted to examine the possibility of detecting different types of lung lesions, such as cancer, using ultra-low dose (ULD) chest CT images.
Method: In this basic (experimental) study with computed tomography (CT images), 20 patients with different lung disease indications were scanned with ULD and routine dose chest CT protocols. ULD and routine dose CT images were reconstructed utilizing iDose and iterative model reconstruction. CT images were evaluated by two expert radiologists. Volume CT dose index (CTDIvol), dose length product, and effective dose were used for dose assessment in both protocols.
Results: CTDIvol and dose length product for ULD protocol were 98% less compared to those for routine chest CT. The chest CT images for ULD and routine dose were diagnosed as normal in three patients with lung lesions, such as nodules, masses, plural effusion, fibrosis, diffuse ground glass opacities, bronchiectasis, and infiltration, in 17 patients. Patient dose of ULD chest CT (0.11mSv) is comparable to Poster-Anterior plus Lateral (0.1 mSv) chest radiograph while the effective dose due to routine chest CT is about 5.1 mSv.
Conclusion: Diagnostic findings regarding ULD chest CT images with 98% of dose reduction were compared to those for routine dose. We concluded that it may be utilized as a very useful tool for screening and the follow-up of different lung diseases, malignancy for instance. ULD chest CT with 98% of dose reduction could be a suitable substitute for chest radiograph, with higher diagnostic values.
Poya Iranpour, Nasrin Namdari, Mehrosadat Alavi, and Bita Geramizadeh
Elsevier BV
BACKGROUND
Cystic angiomatosis is a rare benign disease presents with multiple lytic and sclerotic bone lesions mimicking a metastatic malignant neoplasia with less than 50 cases have been reported in literature so far.
CASE PRESENTATION
We reported a case of a 48-year-old woman who presented to an oncology clinic with multiple lytic and sclerotic bone lesions. Oncologic investigation for metastatic malignant neoplasia started. After that the negative results were obtained by evaluating the primary tumor site, a final diagnosis of cystic angiomatosis was made according to bone biopsy results.
CONCLUSIONS
Cystic angiomatosis is a rare disease with unpredictable prognosis. It can mimic metastatic malignancy especially when it presents at old age.
M. Mokhtari, P. Iranpour, Ardalan Golbahar Haghighi and L. Ghahramani
Schwannoma is a rare tumor in the colon which originates from the peripheral nerve plexus. Most of the cases have been asymptomatic but occasionally present as an obstructive mass. Abdominal investigations are effective in some cases, but usually, they are not informative. A significant number of cases have been detected after their operation by histopathology examination. Immune and histochemical staining shows the spindle cells that have been positive for S-100 and vimentin, but negative for CD34 and smooth muscle actin. If the diagnosis of Schowannoma is confirmed preoperatively, segmental resection is recommended. In this case report, we presented a 58-year-old woman with pelvic mass and normal colonoscopy that mimic extramural large uterine myoma with extraluminal pressure effect on the rectosigmoid.
Kamyar Iravani, Davood Mehrabani, Aida Doostkam, Negar Azarpira, Pooya Iranpour, Mohsen Bahador, and Soheila Mehravar
Elsevier BV
Fariba Zarei, Reza Jalli, Pooya Iranpour, Sepideh Sefidbakht, Sahar Soltanabadi, Maryam Rezaee, Reza Jahankhah, and Alireza Manafi
Elsevier BV
Objectives
To investigate the chest CT and clinical characteristics of COVID-19 pneumonia and H1N1 influenza, and explore the radiologist diagnosis differences between COVID-19 and influenza.
Materials and methods
This cross-sectional study included a total of 43 COVID-19-confirmed patients (24 men and 19 women, 49.90 ± 18.70 years) and 41 influenza-confirmed patients (17 men and 24 women, 61.53 ± 19.50 years). Afterwards, the chest CT findings were recorded and three radiologists recorded their diagnoses of COVID-19 or of H1N1 influenza based on the CT findings.
Results
The most frequent clinical symptom in patients with COVID-19 and H1N1 pneumonia were dyspnea (96.6%) and cough (62.5%), respectively. The CT findings showed that the COVID-19 group was characterized by GGO (88.1%), while the influenza group had features such as GGO (68.4%) and consolidation (66.7%). Compared with the influenza group, the COVID-19 group was more likely to have GGO (88.1% vs. 68.4%, p = 0.032), subpleural sparing (69.0% vs. 7.7%, p <0.001) and subpleural band (50.0% vs. 20.5%, p = 0.006), but less likely to have pleural effusion (4.8% vs. 33.3%, p = 0.001). The agreement rate between the three radiologists was 65.8%.
Conclusion
Considering similarities of respiratory infections especially H1N1 and COVID-19, it is essential to introduce some clinical and para clinical modalities to help differentiating them. In our study we extracted some lung CT scan findings from patients suspected to COVID-19 as a newly diagnosed infection comparing with influenza pneumonia patients.
Sepideh Sefidbakht, Mehrdad Askarian, Bijan Bijan, Mohammad Eghtedari, Sedigheh Tahmasebi, Fariba Zarei, Reza Jalli, and Pooya Iranpour
Elsevier BV
As the Coronavirus disease 2019 (COVID-19) epidemic begins to stabilize, different medical imaging facilities not directly involved in the COVID-19 epidemic face the dilemma of how to return to regular operation. We hereby discuss various fields of concern in resuming breast imaging services. We examine the concerns for resuming functions of breast imaging services in 2 broad categories, including safety aspects of operating a breast clinic and addressing potential modifications needed in managing common clinical scenarios in the COVID-19 aftermath. Using a stepwise approach in harmony with the relative states of the epidemic, health care system capacity, and the current state of performing breast surgeries (and in compliance with the recommended surgical guidelines) can ensure avoiding pointless procedures and ensure a smooth transition to a fully operational breast imaging facility.