Differential privacy preserved federated learning for prognostic modeling in COVID-19 patients using large multi-institutional chest CT dataset.

BACKGROUND: Notwithstanding 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. PURPOSE: This study aimed to evaluate the performance of deep privacy-preserving federated learning (DPFL) in predicting COVID-19 outcomes using chest CT images. METHODS: After 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. RESULTS: The 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. CONCLUSION: The 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.

Influenza Vaccine Booster Stimulates Antibody Response in Beta Thalassemia Major Patients.

The aim of this study was to evaluate antibody response against influenza vaccine in beta thalassemia major patients from Iran. Thirty beta thalassemia major patients were enrolled and divided into three groups: single dose (group 1), double dose (group 2), and control (group 3). Seroconversion, seroprotection, and geometric mean titer (GMT) assays were performed through hemagglutination inhibition (HI) on days 0, 14, and 60. Based on the results, the level of antibody titer was increased in group 2. Two weeks after vaccination, seroconversion rate was about 20% and 30% in groups 1 and 2. Sixty days after vaccination, the seroconversion rate was around 70% and GMT showed a more than 2-fold increase in group 2. Based on the results, the immunogenicity of double dose vaccination against influenza infection appears to be higher than the single dose vaccine in beta thalassemia major patients, and thus it is recommended to use two doses of vaccine, especially in splenectomized patients who are more sensitive than others.

COVID-19 prognostic modeling using CT radiomic features and machine learning algorithms: Analysis of a multi-institutional dataset of 14,339 patients.

BACKGROUND: We aimed to analyze the prognostic power of CT-based radiomics models using data of 14,339 COVID-19 patients. METHODS: Whole lung segmentations were performed automatically using a deep learning-based model to extract 107 intensity and texture radiomics features. We used four feature selection algorithms and seven classifiers. We evaluated the models using ten different splitting and cross-validation strategies, including non-harmonized and ComBat-harmonized datasets. The sensitivity, specificity, and area under the receiver operating characteristic curve (AUC) were reported. RESULTS: In the test dataset (4,301) consisting of CT and/or RT-PCR positive cases, AUC, sensitivity, and specificity of 0.83 ± 0.01 (CI95%: 0.81-0.85), 0.81, and 0.72, respectively, were obtained by ANOVA feature selector + Random Forest (RF) classifier. Similar results were achieved in RT-PCR-only positive test sets (3,644). In ComBat harmonized dataset, Relief feature selector + RF classifier resulted in the highest performance of AUC, reaching 0.83 ± 0.01 (CI95%: 0.81-0.85), with a sensitivity and specificity of 0.77 and 0.74, respectively. ComBat harmonization did not depict statistically significant improvement compared to a non-harmonized dataset. In leave-one-center-out, the combination of ANOVA feature selector and RF classifier resulted in the highest performance. CONCLUSION: Lung CT radiomics features can be used for robust prognostic modeling of COVID-19. The predictive power of the proposed CT radiomics model is more reliable when using a large multicentric heterogeneous dataset, and may be used prospectively in clinical setting to manage COVID-19 patients.

The Immune Responses against Coronavirus Infections: Friend or Foe?

Coronaviruses (CoVs) were first discovered in the 1960s. Severe acute respiratory syndrome CoV-2 (SARS-CoV-2) has been identified as the cause of COVID-19, which spread throughout China and subsequently, across the world. As COVID-19 causes serious public health concerns across the world, investigating the characteristics of SARS-CoV-2 and its interaction with the host immune responses may provide a clearer picture of how the pathogen causes disease in some individuals. Interestingly, SARS-CoV-2 has 80% sequence homology with SARS-CoV-1 and 96-98% homology with CoVs isolated from bats. Therefore, the experience acquired in SARS and Middle East Respiratory Syndrome (MERS) epidemics may improve our understanding of the immune response and immunopathological changes in COVID-19 patients. In the present paper, we have reviewed the immune responses (including the innate and adaptive immunities) to SARS-CoV, MERS-CoV, and SARS-CoV-2, so as to improve our understanding of the concept of the COVID-19 disease, which will be helpful in developing vaccines and medications for treating the COVID-19 patients.

Molecular imaging in tracking cancer stem cells: A review.

Cancer stem cells (CSCs) have critical roles in tumor development, progression, and recurrence. They are responsible for current cancer treatment failure and remain questionable for the design and development of new therapeutic strategies. With this issue, medical imaging provides several clues for finding biological mechanisms and strategies to treat CSCs. This review aims to summarize current molecular imaging approaches for detecting CSCs. In addition, some promising issues for CSCs finding and explaining biological mechanisms have been addressed. Among the molecular imaging approaches, modalities including Magnetic resonance imaging (MRI) and positron emission tomography (PET) have the greatest roles and several new approaches such as optical imaging are in progress.

Tumor cryotherapy using Ice-producing bacteria.

One of he approaches to cancer treatment is cryotherapy. In this therapy low temperatures lead to freezing and killing the cancer cells. Low temperature has several side effects on the health of tissues. Using bacteria for treatment of cancer as a therapeutic approach is proposed. One of the bacteria is Pseudomonas syringe with ice-producing properties. In this study, we hypothesized that by insertion of INA gene of P. syringe into anaerobic bacteria can do cryotherapy at a low temperature. This hypothesis is based on the manipulated anaerobic bacteria moves to the side of the tumor from ice crystal.

Cancer stem cells: A review from origin to therapeutic implications.

Cancer stem cells (CSCs), also known as tumor-initiating cells (TICs), are elucidated as cells that can perpetuate themselves via autorestoration. These cells are highly resistant to current therapeutic approaches and are the main reason for cancer recurrence. Radiotherapy has made a lot of contributions to cancer treatment. However, despite continuous achievements, therapy resistance and tumor recurrence are still prevalent in most patients. This resistance might be partly related to the existence of CSCs. In the present study, recent advances in the investigation of different biological properties of CSCs, such as their origin, markers, characteristics, and targeting have been reviewed. We have also focused our discussion on radioresistance and adaptive responses of CSCs and their related extrinsic and intrinsic influential factors. In summary, we suggest CSCs as the prime therapeutic target for cancer treatment.

Pyrin and Hematopoietic Interferon-Inducible Nuclear Protein Domain Proteins: Innate Immune Sensors for Cytosolic and Nuclear DNA.

The innate immune system is the first line of defense against microbial pathogens. The response of innate immunity is initiated by molecules known as pattern recognition receptors (PRRs). Such responses are often triggered by nucleic acids that are delivered to the cytoplasm or nucleus of cells. The ability to recognize foreign nucleic acids in these two locations is an important defense mechanism of the human innate immune system. Several PRRs are located in the cytosol or nucleus and detect foreign DNAs. The pyrin and hematopoietic interferon-inducible nuclear (PYHIN) domain protein is a family of PRRs that includes interferon-inducible protein 16, absent in melanoma 2, PYHIN 1 (or interferon-inducible protein X, as it is also known), myeloid cell nuclear differentiation antigen, and pyrin domain only protein 3. These nuclear and cytosolic sensors play an essential part in host defense of intracellular pathogens. In addition, members of the PYHIN family are critical regulators of immune response, apoptosis, cell growth, differentiation, and transcription. In this review, we summarize important characteristics of these innate immune sensors and their roles in several diseases. A better understanding of the role of DNA sensors in the nucleus and cytoplasm will lead to the development of novel therapeutic approaches to control infections and associated diseases.

Regulatory Effects of Estradiol on Peripheral Blood Mononuclear Cells Activation in Patients with Asthma.

Asthma prevalence and severity are greater in women than in men, and mounting evidence suggests this is in part related to female steroid sex hormones. Conflicting data are reported regarding pro- and anti-inflammatory properties of estradiol. This study was designed to clarify whether estradiol may contribute to enhanced T helper (Th) 17-associated cytokines production by peripheral blood mononuclear cells (PBMC) in asthmatic patients and healthy individuals. PBMCs from patients with asthma and healthy donors were cultured with 17-ß estradiol (E2) and phytohemagglutinin (PHA). The quantitative real-time polymerase chain reaction (qRT-PCR) was used to measure IL-6, IL-17, IL-23 and TGF-ß. We observed a significant increased IL-17, IL-23 and TGF-ß expression in PBMCs of patients compared to the healthy individuals. In addition, our findings indicated that IL-6 and IL-17 expressions in PBMCs were induced, following E2 treatment. Our results identified an impact of E2 in stimulation of Th17 phenotype, and upon hormonal oscillations and hormone replacement therapy (HRT), asthma inflammation may be mediated by Th17-associated cytokines.

Human adipose tissue-derived mesenchymal stem cells in rheumatoid arthritis: Regulatory effects on peripheral blood mononuclear cells activation.

BACKGROUND AND OBJECTIVES: Mesenchymal stem cells (MSCs) are multipotent adult stem cells with immunomodulatory properties. The mechanisms by which MSCs inhibit the proliferation of pro-inflammatory T cells have not been fully elucidated yet. It is assumed that pro-inflammatory T-cells play an important role in the development of autoimmune diseases. We investigated the potential therapeutic effects of human adipose tissue derived (Ad)-MSCs on the peripheral blood mononuclear cells (PBMCs) of rheumatoid arthritis (RA) patients and healthy individuals, with a particular focus on Th17-associated cytokines. MATERIALS AND METHODS: PBMCs from RA patients and healthy donors were co-cultured with Ad-MSCs and HeLa with or without Phytohemagglutinin (PHA). Finally, IL-6, IL-17, IL-21, IL-23 and TGF-ß levels were determined by ELISA and quantitative real-time RT-PCR on co-culture supernatants and PBMCs, respectively. RESULTS: In co-culture interaction, Ad-MSCs inhibited IL-17 secretion by PBMCs compared to unstimulated PBMCs cultured alone. In addition, IL-21 expressions in PBMCs of the patient group, and IL-17 and IL-21 in healthy group were inhibited by Ad-MSCs compared to PBMCs cultured alone. TGF-ß expression in healthy individuals remarkably increased in both MSC-treated groups with and without PHA in comparison to PHA-stimulated and -unstimulated PBMCs. CONCLUSIONS: This study demonstrates that human Ad-MSCs act as key regulators of immune tolerance by inhibiting the inflammation. Therefore, they can be attractive candidates for immunomodulatory cell-based therapy in RA.

Elevated IL-17 and TGF-ß Serum Levels: A Positive Correlation between T-helper 17 Cell-Related Pro-Inflammatory Responses with Major Depressive Disorder.

INTRODUCTION: Depression is a mental disorder that highly associated with immune system. Therefore, this study compares the serum concentrations of IL-21, IL-17, and transforming growth factor ß (TGF-ß) between patients with major depressive disorder and healthy controls. METHODS: Blood samples were collected from 41 patients with major depressive disorder and 40 healthy age-matched controls with no history of malignancies or autoimmune disorders. The subjects were interviewed face to face according to DSM-IV diagnostic criteria. Depression score was measured using completed Beck Depression Inventory in both groups. The serum concentrations of IL-21, IL-17, and TGF-ß were assessed using ELISA. RESULTS: The mean score of Beck Depression score in the patient and control groups was 35.4±5.5 and 11.1±2.3. IL-17 serum concentrations in the patients and the control group were 10.03±0.6 and 7.6±0.6 pg/mL, respectively (P=0.0002). TGF-ß level in the patients group was significantly higher than compare to the control group; 336.7±20.19 vs. 174.8±27.20 pg/mL, (P<0.0001). However, the level of IL-21 was not statistically different between the two groups 84.30±4.57 vs. 84.12±4.15 pg/mL (P>0.05). CONCLUSION: Considering pro-inflammatory cytokines, current results support the association of inflammatory response and depressive disorder. So, it seems that pro-inflammatory factors profile can be used as indicator in following of depression progress and its treatment impacts.

Reduced levels of T-helper 17-associated cytokines in the serum of patients with breast cancer: indicators for following the course of disease.

Interleukin (IL)-17-producing CD4(+) T helper (Th17) cells that are known to produce IL-17 have recently been defined as a unique subset of proinflammatory helper cells. Interleukin 17 is an inflammatory cytokine with robust effects on many cells. It can play important roles in the pathogenesis of diverse groups of immune-mediated diseases. In this regard, the present case-control study aimed at determining serum levels of IL-17, IL-6, and transforming growth factor ß (TGF-ß) in Iranian breast cancer patients. Blood samples were collected from 55 patients with breast cancer and 34 healthy individuals with no history of malignancies or autoimmune disorders, based on simple sampling. The serum levels of IL-17, IL-6 and TGF-ß were measured by enzyme-linked immunosorbent assay (ELISA). The serum level of IL-6 was significantly lower in patients with breast cancer compared with healthy individuals (p = 0.0003), and also the IL-17 was lower in the patient group than in controls (p = 0.01). Interestingly, the TGF-ß serum level in patients was less than in controls (p < 0.0001). As most of the cases investigated in this study were in their early stages, it can be concluded that reduced IL-17, IL-6, and TGF-ß can be used as predictors for clinical stage and prognosis of cancers such as breast carcinoma.