Lung cancer detection based on computed tomography image using convolutional neural networks

BACKGROUND

Lung cancer is the most common type of cancer, accounting for 12.8% of cancer cases worldwide. As initially non-specific symptoms occur, it is difficult to diagnose in the early stages.

OBJECTIVE

Image processing techniques developed using machine learning methods have played a crucial role in the development of decision support systems. This study aimed to classify benign and malignant lung lesions with a deep learning approach and convolutional neural networks (CNNs).

METHODS

The image dataset includes 4459 Computed tomography (CT) scans (benign, 2242; malignant, 2217). The research type was retrospective; the case-control analysis. A method based on GoogLeNet architecture, which is one of the deep learning approaches, was used to make maximum inference on images and minimize manual control.

RESULTS

The dataset used to develop the CNNs model is included in the training (3567) and testing (892) datasets. The model's highest accuracy rate in the training phase was estimated as 0.98. According to accuracy, sensitivity, specificity, positive predictive value, and negative predictive values of testing data, the highest classification performance ratio was positive predictive value with 0.984.

CONCLUSION

The deep learning methods are beneficial in the diagnosis and classification of lung cancer through computed tomography images.

Deep learning for diagnosis of malign pleural effusion on computed tomography images

BACKGROUND

The pleura is a serous membrane that surrounds the lungs. The visceral surface secretes fluid into the serous cavity and the parietal surface ensures a regular absorption of this fluid. If this balance is disturbed, fluid accumulation occurs in the pleural space called "Pleural Effusion". Today, accurate diagnosis of pleural diseases is becoming more critical, as advances in treatment protocols have contributed positively to prognosis. Our aim is to perform computer-aided numerical analysis of Computed Tomography (CT) images from patients showing pleural effusion images on CT and to examine the prediction of malignant/benign distinction using deep learning by comparing with the cytology results.

METHODS

The authors classified 408 CT images from 64 patients whose etiology of pleural effusion was investigated using the deep learning method. 378 of the images were used for the training of the system; 15 malignant and 15 benign CT images, which were not included in the training group, were used as the test.

RESULTS

Among the 30 test images evaluated in the system; 14 of 15 malignant patients and 13 of 15 benign patients were estimated with correct diagnosis (PPD: 93.3%, NPD: 86.67%, Sensitivity: 87.5%, Specificity: 92.86%).

CONCLUSION

Advances in computer-aided diagnostic analysis of CT images and obtaining a pre-diagnosis of pleural fluid may reduce the need for interventional procedures by guiding physicians about which patients may have malignancies. Thus, it is cost and time-saving in patient management, allowing earlier diagnosis and treatment.

Distant Metastasis Patterns of Lung Cancer on Positron Emission Tomography/Computed Tomography Association with Age and Histological Subtype

OBJECTIVE

To investigate whether age or other factors are determinants of distant metastasis in patients with lung cancer.

STUDY DESIGN

Observational study.

PLACE AND DURATION OF STUDY

Department of Nuclear Medicine, Recep Tayyip Erdogan University, Rize, Turkey between December, 2018 and February, 2019.

METHODOLOGY

A total of 152 patients with lung cancer, who underwent positron emission tomography/computed tomography (PET/CT) for staging, were included in this study. Patients were grouped according to age (>65 and <65 years) and distant metastasis status. Metastasis localisation of patients was evaluated by PET/CT. Univariate/multivariate regression analyses were performed to determine risk factors for distant metastasis.

RESULTS

No significant difference was found when the relation of distant metastasis with stage distribution was examined in both age groups. Distant metastasis rates were significantly higher in female patients than in male patients (p = 0.019) and in patients with small-cell lung carcinoma (SCLC)-adenocarcinoma than in those with other histopathological subtypes (p <0.001). Most of the patients in both groups had a stage 4 disease, and bone distant metastasis was the most common in both age groups. Univariate/multivariate analyses identified that female gender (p = 0.017/p = 0.003), SCLC subtype (p = 0.013/p = 0.008), T3/T4 tumour (p <0.001/p <0.001), and smoking history of >66 pack-years (p = 0.047/p = 0.047) were independent factors for the presence of distant metastasis.

CONCLUSION

Although age is not a risk factor for distant metastasis in lung cancer, female gender, T3/4 tumour, SCLC subtype, and smoking history of >66 pack-years are high-risk factors. PET/CT is recommended as the first-choice imaging technique in patients with lung cancer indicated for distant metastasis scanning. Key Word: Lung cancer, PET/CT, Metastases, Histological subtype, T-stage.

The Role of Hematological Inflammatory Biomarkers in the Diagnosis of Lung Cancer and in Predicting TNM Stage

OBJECTIVE

The aim of study is to investigate whether hematological inflammatory biomarkers could be useful to detect patients with lung cancer.

METHODS

The contribution of hematological biomarkers to the diagnosis of lung cancer and prediction of TNM was examined.

RESULTS

NLR, PLR, MPV values were found to be higher in patients with lung cancer (all p < .001). NLR and PLR were found to be high, MPV was found to be lower in disease of advanced stage (p < .001).

CONCLUSIONS

This study found that NLR, PLR and MPV values were significantly higher in patients with lung cancer.

Effect of coronaphobia on smoking habits

INTRODUCTION AND OBJECTIVE

Coronavirus disease 2019 (COVID-19) epidemic had exhibited negative effects on patients with a nicotine use disorder because of social stress situations. The objective of this study is to investigate the changes that occurred in the smoking habits of the patients who applied to our smoking cessation clinic because of the COVID-19 outbreak's effect (coronaphobia) on their mental health.

METHOD

In this study, we included the patients who applied to our smoking cessation clinic between January 1 and June 30, 2020. We conducted telephonic and face-to-face interviews with the patients during the outbreak, and inquired about the changes in their smoking habits.

RESULTS

In total, 114 volunteers (mean age: 41.2 years [min-max: 18-82 years], M/F: 70/44 [61.4/38.6%]) who registered in the smoking cessation clinic of Chest Diseases Clinic were included in the study. The number of people with and without coronaphobia was 64 (56.1%) and 50 (43.9%), respectively. The analyses of the changes in smoking behaviors because of coronaphobia demonstrated that people with coronaphobia exhibited a significantly higher decrease or cessation of smoking as compared to no change in smoking behavior or increase in the number of cigarettes (p: .006).

CONCLUSIONS

Success in smoking cessation can be increased with the accurate analyses of the anxiety effects caused by coronaphobia after COVID-19 outbreak on the desire of smoking cessation. This resolution depends on the solution-oriented strategies that will increase the awareness of physicians and patients, a close follow-up of patients with tools such as telemedicine, and a quick determination of their needs.

Can artificial intelligence distinguish between malignant and benign mediastinal lymph nodes using sonographic features on EBUS images?

AIMS

This study aimed to develop a new intelligent diagnostic approach using an artificial neural network (ANN). Moreover, we investigated whether the learning-method-guided quantitative analysis approach adequately described mediastinal lymphadenopathies on endobronchial ultrasound (EBUS) images.

METHODS

In total, 345 lymph nodes (LNs) from 345 EBUS images were used as source input datasets for the application group. The group consisted of 300 and 45 textural patterns as input and output variables, respectively. The input and output datasets were processed using MATLAB. All these datasets were utilized for the training and testing of the ANN.

RESULTS

The best diagnostic accuracy was 82% of that obtained from the textural patterns of the LNs pattern (89% sensitivity, 72% specificity, and 78.2% area under the curve). The negative predictive values were 81% compared to the corresponding positive predictive values of 83%. Due to the application group's pattern-based evaluation, the LN pattern was statistically significant (p = .002).

CONCLUSIONS

The proposed intelligent approach could be useful in making diagnoses. Further development is required to improve the diagnostic accuracy of the visual interpretation.

The value of the platelet count and platelet indices in differentiation of COVID-19 and influenza pneumonia

It is difficult to distinguish coronavirus disease-2019 (COVID-19) from other viral respiratory tract infections owing to the similarities in clinical and radiological findings. This study aims to determine the clinical importance of platelet count and platelet indices in the differentiation of COVID-19 from influenza and the value of these parameters in the differential diagnosis of COVID-19. The medical records of the patients and the electronic patient monitoring system were retrospectively analyzed. Demographic characteristics, admission symptoms, laboratory findings, radiological involvement, comorbidities, and mortality of the patients were recorded. Forty-three patients diagnosed with influenza and 54 diagnosed with COVID-19 were included in the study. The average age of the COVID-19 patients was lower than that of the influenza patients (influenza: 60.5 years, COVID-19: 52.4 years; pp = 0.024),.024), and the male gender was predominant in the COVID-19 group (influenza: 42%, COVID-19: 56%). According to laboratory findings, the mean platelet volume (MPV) and MPV/platelet ratio were statistically significantly lower, whereas the eosinophil count and platelet distribution width levels were significantly higher (p < 0.05) in the COVID-19 group. It was found that the most common symptom in both groups was dyspnea and that the symptom was more prevalent among influenza patients. In the diagnosis of COVID-19, the platelet count and platelet indices are easily accessible, inexpensive, and important parameters in terms of differential diagnosis and can help in the differentiation of COVID-19 from influenza during seasonal outbreaks of the latter.

Unclear Issues Regarding COVID-19

Scientists from all over the world have been intensively working to discover different aspects of Coronavirus disease 2019 (COVID-19) since the first cluster of cases was reported in China. Herein, we aimed to investigate unclear issues related to transmission and pathogenesis of disease as well as accuracy of diagnostic tests and treatment modalities. A literature search on PubMed, Ovid, and EMBASE databases was conducted, and articles pertinent to identified search terms were extracted. A snow-ball search strategy was followed in order to retrieve additional relevant articles. It was reported that viral spread may occur during the asymptomatic phase of infection, and viral load was suggested to be a useful marker to assess disease severity. In contrast to immune response against viral infections, cytotoxic T lymphocytes decline in SARS-CoV-2 infection, which can be partially explained by direct invasion of T lymphocytes or apoptosis activated by SARS-CoV-2. Dysregulation of the urokinase pathway, cleavage of the SARS-CoV-2 Spike protein by FXa and FIIa, and consumption coagulopathy were the proposed mechanisms of the coagulation dysfunction in COVID-19. False-negative rates of reverse transcriptase polymerase chain reaction varied between 3% and 41% across studies. The probability of the positive test was proposed to decrease with the number of days past from symptom onset. Safety issues related to infection spread limit the use of high flow nasal oxygen (HFNO) and continuous positive airway pressure (CPAP) in hypoxic patients. Further studies are required to elucidate the challenging issues, thus enhancing the management of COVID-19 patients.

Prevalence of non-thyroidal illness syndrome in COPD exacerbation and effect of hypoxaemia and hypercapnia on thyroid functions

INTRODUCTION

Chronic obstructive pulmonary disease (COPD) is a chronic inflammatory disease showing acute exacerbations during its course. Comorbidities often accompany. Non-thyroidal illness syndrome (NTIS) occurs because of the functional impairment in the hypothalamic-pituitary-thyroid axis in severe critical cases. The objective of the current study is to determine the prevalence of NTIS among hospitalised patients due to acute exacerbation of chronic obstructive pulmonary disease (AECOPD) and to reveal the factors affecting thyroid functions.

MATERIALS AND METHODS

A total of 132 patients hospitalised for AECOPD were enroled. Arterial blood gas samples at room air and venous blood samples for thyroid function tests were obtained within 24 hours following hospitalisation.

RESULTS

The mean age was 69.3 ± 9.6 years with male predominance (M/F:130/2). The prevalence of NTIS was 55%. Subgroups of NTIS cases were analysed. Low TSH levels were the most common pathology (55%). Patients with NTIS had significantly lower PaO₂ and SaO₂ levels compared with those without NTIS (P = 0.045 and P = 0.022, respectively). In addition, a positive correlation was found between PaO₂ and free thyroxine (fT4) (P < 0.001, r = 0.313). A statistically significant negative correlation between PaCO₂ and fT4 was found (P < 0.001, r = -0.393). And a statistically significant negative correlation between PaCO₂ and free triiodothyronine (fT3) values were found (P = 0.040, r = -0.183).

CONCLUSION

NTIS is a very common condition during AECOPD. We believe that hypoxemia causing functional impairment in the hypothalamic-pituitary-thyroid axis is the main mechanism in NTIS development and hypercapnia disrupts freeT3 and freeT4 production and secretion.

Assessment of Palliative Care in Lung Cancer in Turkey

OBJECTIVE

To investigate the symptoms of lung cancer in Turkey and to evaluate approaches to alleviate these symptoms.

SUBJECTS AND METHODS

This study included 1,245 lung cancer patients from 26 centers in Turkey. Demographic characteristics as well as information regarding the disease and treatments were obtained from medical records and patient interviews. Symptoms were evaluated using the Edmonton Symptom Assessment Scale (ESAS) and were graded on a scale between 0 and 10 points. Data were compared using the χ2, Student t, and Mann-Whitney U tests. Potential predictors of symptoms were analyzed using logistic regression analysis.

RESULTS

The most common symptom was tiredness (n = 1,002; 82.1%), followed by dyspnea (n = 845; 69.3%), appetite loss (n = 801; 65.7%), pain (n = 798; 65.4%), drowsiness (n = 742; 60.8%), anxiety (n = 704; 57.7%), depression (n = 623; 51.1%), and nausea (n = 557; 45.5%). Of the 1,245 patients, 590 (48.4%) had difficulty in initiating or maintaining sleep. The symptoms were more severe in stages III and IV. Logistic regression analysis indicated a clear association between demographic characteristics and symptom distress, as well as between symptom distress (except nausea) and well-being. Overall, 804 (65.4%) patients used analgesics, 630 (51.5%) received treatment for dyspnea, 242 (19.8%) used enteral/parenteral nutrition, 132 (10.8%) used appetite stimulants, and 129 (10.6%) used anxiolytics/antidepressants. Of the 799 patients who received analgesics, 173 (21.7%) reported that their symptoms were under control, and also those on other various treatment modalities (dyspnea: 78/627 [12.4%], appetite stimulant: 25/132 [18.9%], and anxiolytics/antidepressants: 25/129 [19.4%]) reported that their symptoms were controlled.

CONCLUSION

In this study, the symptoms progressed and became more severe in the advanced stages of lung cancer, and palliative treatment was insufficient in most of the patients in Turkey.

Prognostic value of red cell distribution width in patients with sarcoidosis

BACKGROUND

Sarcoidosis is a multisystem granulomatous disorder of unknown etiology. There are no data on the indications for specific tests or optimal frequency for monitoring of the disease activity in sarcoidosis.

METHODS

Complete blood counts, demographics and pulmonary function data from sarcoidosis patients evaluated between 2006 and 2012 were collated retrospectively. During follow-up, the latest red cell distribution width (RDW) values of the patients were recorded. The prognosis and diagnosis of sarcoidosis was based on according to the guideline. Sarcoidosis progression was classified as follows: remission, stable disease and progreesive disease.

RESULTS

The diagnosis was based on histopathological findings in 93 out of 138 (67.4%) patients. In our cohort, the baseline mean RDW levels were 14.1% ± 1.2. The RDW distribution by stage was as follows: stage I: 14.0% (± 1.6), II: 14.2% (± 1.5), III: 13.8% (± 0.9) and IV: 15.8% (± 2.0). In patients with stage IV, baseline and follow-up values of RDW were found to be significantly higher than the other stages. While the mean baseline RDW was 14.8 (± 1.4) in the progressive disease, upon follow-up, the mean RDW had increased to 16.5% (± 1.4) (P = 0.021), No difference was found between the baseline and follow-up levels of RDW in the regressive and stable groups.

CONCLUSIONS

Serial RDW levels may be beneficial marker to predict progression of sarcoidosis.

The protective effect of caffeic acid phenethyl ester (CAPE) on oxidative stress in rat liver exposed to the 900 MHz electromagnetic field

In this study, we aimed to investigate the possible protective effects of caffeic acid phenethyl ester (CAPE) on lipid peroxidation (LPO) and the activities of antioxidant enzymes in the liver of rats exposed to the 900 MHz electromagnetic field (EMF). EMF of cellular phones may affect biological systems by increasing free radical, which appear mainly to enhance LPO, and by changing the antioxidative activities of liver, thus leading to oxidative damage. CAPE, an active component of propolis extract, exhibits antioxidant properties and several studies suggest that supplementation with antioxidant can influence EMF exposure induced hepatotoxicity. Thirty male Sprague-Dawley rats were divided into three groups: control ( n = 10), 900 MHz EMF ( n = 10) and 900 MHz EMF + CAPE ( n = 10). CAPE was injected intraperitoneally for 30 days before exposure to EMF. Liver tissue was removed to study the activities of catalase (CAT), superoxide dismutase (SOD), glutathione peroxidase (GSH-Px), xanthine oxidase (XO) and the levels of LPO. The activities of XO, CAT and level of LPO increased in the 900 MHz electromagnetic field (EMF) group compared with the control group, although XO, CAT activities and LPO levels were decreased by 900 MHz EMF + CAPE administration. The activities of SOD and GSH-Px decreased in the 900 MHz EMF group compared with the control group, although their levels were increased by EMF + CAPE administration. It can be concluded that CAPE may prevent the 900 MHz EMF-induced oxidative changes in liver by strengthening the antioxidant defense system by reducing reactive oxygen species and increasing antioxidant enzyme activities.

Myocardial perfusion is preserved in patients with psoriasis without clinically evident cardiovascular disease

BACKGROUND

Psoriasis is associated with a premature atherosclerosis due to the chronic inflammatory process. To evaluate the effect of disease process on myocardial perfusion, we planned to perform 99mTc-MIBI myocardial perfusion single photon emission computed tomography (SPECT) in patients with psoriasis.

METHODS

The study group consisted of 28 psoriasis patients (17 men, 11 women), aged 18-76 years, and mean age 41.2 +/- 14.1 years. The patients were selected among those who were older than 18 years and longer than 10 years of disease duration with more than two times of systemic treatment. All patients underwent 99mTc-MIBI myocardial perfusion SPECT with the same day protocol.

RESULTS

We detected various risk factors including smoking habits in 7, family history of cardiovascular disease in 4, hypertension in 1, hyperlipidemia in 9 patients. We completed myocardial perfusion SPECT for each patient and found normal perfusion pattern in SPECT images.

CONCLUSION

We detected that myocardial perfusion is preserved in the patients with psoriasis. The majority of acute heart attacks are caused by noncritical coronary stenosis and this may be an explanation for increased cardiovascular risk in these patients despite normal coronary perfusion.

Prevalence of interatrial septal aneurysm in newborns and their natural course

The objective of this study was to evaluate the prevalence of atrial septal aneurysm (ASA) in newborns, to define the natural course of ASA, and to investigate its role on closure of associated interatrial septal opening (IASO). A total of 1072 consecutive neonates were examined with echocardiography in the early postnatal period. The length of the interatrial septum, the diameter of IASO, the excursion and base of aneurysm, and the width of the related atrium were measured and the excursion ratio and the basal ratio were calculated for each neonate. Aneurysms with an excursion ratio > or = 25% were diagnosed as ASA. There were 81 neonates (7.6%) with ASA. The prevalence of ASA was 11.1% in preterm (14 of 126) and 7.1% in full-term newborns (67 of 946). All of the ASAs disappeared at the end of the first year of life, and there were no complications related to the lesion during the follow-up period. Although overall IASO prevalence was 78.6% (843 of 1072), it was 72.8% (59 of 81) among the cases with ASA. Although the disappearance time of interatrial septal shunt was not significantly different between the cases with and without ASA, spontaneous closure was less frequent in the cases with ASA than in those without ASA 77.7 and 96.1%), respectively (p < 0.001). The prevalence of ASA is high among newborns, with a high resolution rate. Therefore, it can be considered that it is benign and transient observation. Less frequent spontaneous closure of IASO in cases with ASA indicates that ASA may have a deleterious effect on spontaneous closure.

Toxin production by Alternaria alternata in tomatoes and apples stored under various conditions and quantitation of the toxins by high-performance liquid chromatography

Alternaria alternata strain 8442-3 was inoculated into tomatoes (Lycopersicon esculentum Mill.) and Red Delicious cultivar apples (Malus domestica Borkh.). Half of the lots of each fruit were shrink-wrapped in high-density polyethylene film. Wrapped and unwrapped fruits were incubated under darkness at 4, 15 and 25 degrees C for up to 5 weeks. A high-performance liquid chromatography method was developed to quantitate tenuazonic acid (TeA), alternariol (AOH), alternariol methyl ether (AME), and altenuene (ALT). Shrink-wrapping retarded, but did not completely inhibit growth in tomatoes for 3-7 days. Concentrations of up to 120.6 mg of AOH and 63.7 mg of AME per 100 g of tissue were produced in unwrapped tomatoes stored at 15 degrees C for 4 weeks; 19.0 mg of ALT per 100 g of tomato tissue was produced after 3 weeks at 25 degrees C. AOH, AME and ALT were also produced in unwrapped tomatoes stored at 4 degrees C; however, no TeA was detected in decayed tomatoes, regardless of type of wrapping or storage temperature. Shrink-wrapping resulted in decreased production of AOH, AME, and ALT. Alternaria toxins were not detected in apples stored at 4 and 15 degrees C. The highest concentration of AOH produced (48.8 mg per 100 g of tissue) was in unwrapped apples stored at 25 degrees C for 2 weeks; 12.3 mg per 100 g of tissue of shrink-wrapped apples was detected after 5 weeks of storage at 25 degrees C, while ALT reached 5.7 mg per 100 g after 4 weeks. TeA was not detected in apples infected with A. alternata.