Automated detection and quantification of COVID-19 pneumonia: CT imaging analysis by a deep learning-based software.

BACKGROUND: The novel coronavirus disease 2019 (COVID-19) is an emerging worldwide threat to public health. While chest computed tomography (CT) plays an indispensable role in its diagnosis, the quantification and localization of lesions cannot be accurately assessed manually. We employed deep learning-based software to aid in detection, localization and quantification of COVID-19 pneumonia. METHODS: A total of 2460 RT-PCR tested SARS-CoV-2-positive patients (1250 men and 1210 women; mean age, 57.7 ± 14.0 years (age range, 11-93 years) were retrospectively identified from Huoshenshan Hospital in Wuhan from February 11 to March 16, 2020. Basic clinical characteristics were reviewed. The uAI Intelligent Assistant Analysis System was used to assess the CT scans. RESULTS: CT scans of 2215 patients (90%) showed multiple lesions of which 36 (1%) and 50 patients (2%) had left and right lung infections, respectively (> 50% of each affected lung's volume), while 27 (1%) had total lung infection (> 50% of the total volume of both lungs). Overall, 298 (12%), 778 (32%) and 1300 (53%) patients exhibited pure ground glass opacities (GGOs), GGOs with sub-solid lesions and GGOs with both sub-solid and solid lesions, respectively. Moreover, 2305 (94%) and 71 (3%) patients presented primarily with GGOs and sub-solid lesions, respectively. Elderly patients (≥ 60 years) were more likely to exhibit sub-solid lesions. The generalized linear mixed model showed that the dorsal segment of the right lower lobe was the favoured site of COVID-19 pneumonia. CONCLUSION: Chest CT combined with analysis by the uAI Intelligent Assistant Analysis System can accurately evaluate pneumonia in COVID-19 patients.

Therapeutic effects of sequential therapy by electric coagulation, cryotherapy and balloon dilation with an electronic video bronchoscope.

The aim of the current study was to retrospectively analyze clinical data concerning bronchostenosis or bronchial obstruction caused by endobronchial tuberculosis. Fifty-six cases were subjected to bronchoscopy and chest computed tomography to assess the prognosis of bronchostenosis and bronchial obstruction. Based on reliable and effective anti-pulmonary tuberculosis therapy, these conditions were treated sequentially by electric coagulation, cryotherapy and balloon dilation with an electronic video bronchoscope during outpatient consultation or inpatient hospitalization. Fifty-three subjects with bronchostenosis recovered to varying degrees, a recovery rate of 94.6%. Thirteen of the 15 cases with bronchial obstruction reopened (86.7%). The clinical symptoms of these cases appeared to be in remission. Bronchostenosis or bronchial obstruction resulting from endobronchial tuberculosis may be treated by electric coagulation, cryotherapy and balloon dilation with an electronic video bronchoscope.

Tanshinone IIA attenuates seawater aspiration-induced lung injury by inhibiting macrophage migration inhibitory factor.

Inflammation takes responsibility for the seawater aspiration-induced lung injury. Tanshinone IIA (TIIA) can protect lipopolysaccharide-induced lung injury in mice through the inhibition of inflammation, but it is not reported whether TIIA have a protective effect on lung injury induced by seawater aspiration. Macrophage migration inhibitory factor (MIF) plays an important role in acute lung injury. In this study, we observed the effect of TIIA on the seawater aspiration-induced lung injury and the role of MIF in it. Seawater was aspirated into trachea of rats to make the lung injury model. TIIA was administered to investigate its beneficial effect on seawater-induced acute lung injury. The results showed that seawater aspiration led to hyoxemia, pulmonary edema, neutrophil infiltration, and lung histopathologic changes, with the elevated MIF expression in the lung tissues and plasma. However, these changes were attenuated by TIIA. In macrophage cells we also demonstrated that TIIA could inhibit MIF expression, nuclear factor κB (NF-κB) activity and release of interleukin-6 (IL-6) and tumor necrosis factor-α (TNF-α) induced by seawater. Besides, pretreatment with (S,R)-3-(4-hydroxyphenyl)-4,5-dihydro-5-isoxazole acetic acid (ISO-1), the MIF antagonist, elevated NF-κB and cytokines induced by seawater were also reduced markedly. Furthermore, rMIF treatment alone increased the phosphorylation level of NF-κB and release of cytokines, which was almost abolished by TIIA. Taken together, our results suggested that TIIA exert a protective effect on the seawater aspiration-induced lung injury partly through downregulation of MIF and the subsequent NF-κB activity, as well as expression of IL-6 and TNF-α.

Biomarker discovery and identification from non-small cell lung cancer sera.

Currently, serum biomarkers might usually be thought not to be used for early detection of lung cancer by some researchers. In this study, we used a highly optimized ClinProt-matrix-assisted laser desorption/ionization time-of flight mass spectrometer (MALDI-TOF-MS) to screen non-small cell lung carcinoma (NSCLC) markers in serum. A training set of spectra derived from 45 NSCLC patients, 24 patients with benign lung diseases (BLDs) and 21 healthy individuals, was used to develop a proteomic pattern that discriminated cancer from non-cancer effectively. A test set, including 74 cases (29 NSCLC patients and 45 controls), was used to validate this pattern. After cross-validation, the classifier showed sensitivity and specificity, 86.20% and 80.00%, respectively. Remarkably, 100% of early stage serum samples could be correctly classified as lung cancer. Furthermore, the differential peptides of 1865Da and 4209Da were identified as element of component 3 and eukaryotic peptide chain release factor GTP-binding subunit ERF, respectively. The patterns we described and peptides we identified may have clinical utility as surrogate markers for detection and classification of NSCLC.

Study of differential proteins in lung adenocarcinoma using laser capture microdissection combined with liquid chip-mass spectrometry technology.

BACKGROUND: In recent years the proportion of lung adenocarcinoma (adCA) which occurs in lung cancer patients has increased. Using laser capture microdissection (LCM) combined with liquid chip-mass spectrometry technology, we aimed to screen lung cancer biomarkers by studying the proteins in the tissues of adCA. METHODS: We used LCM and magnetic bead based weak cation exchange (MB-WCX) to separate and purify the homogeneous adCA cells and normal cells from six cases of fresh adCA and matched normal lung tissues. The proteins were analyzed and identified by matrix assisted laser desorption/ionization time-of-fight mass spectrometry (MALDI-OF-MS). We screened for the best pattern using a radial basic function neural network algorithm. RESULTS: About 2.895 × 10(6) and 1.584 × 10(6) cells were satisfactorily obtained by LCM from six cases of fresh lung adCA and matched normal lung tissues, respectively. The homogeneities of cell population were estimated to be over 95% as determined by microscopic visualization. Comparing the differentially expressed proteins between the lung adCA and the matched normal lung group, 221 and 239 protein peaks, respectively, were found in the mass-to-charge ration (M/Z) between 800 Da and 10 000 Da. According to t test, the expression of two protein peaks at 7521.5 M/Z and 5079.3 M/Z had the largest difference between tissues. They were more weakly expressed in the lung adCA compared to the matched normal group. The two protein peaks could accurately separate the lung adCA from the matched normal lung group by the sample distribution chart. A discriminatory pattern which can separate the lung adCA from the matched normal lung tissue consisting of three proteins at 3358.1 M/Z, 5079.3 M/Z and 7521.5 M/Z was established by a radial basic function neural network algorithm with a sensitivity of 100% and a specificity of 100%. CONCLUSIONS: Differential proteins in lung adCA were screened using LCM combined with liquid chip-mass spectrometry technology, and a biomarker model was established. It is possible that this technology is going to become a powerful tool in screening and early diagnosis of lung adCA.

[The application of combined interventional procedures for the management of intractable central airway stenosis].

OBJECTIVE: to evaluate the application of combined five interventional procedures in the management of intractable central airway stenosis. METHODS: clinical manifestations and pulmonary functions of 138 patients with intractable central airway stenosis were evaluated. Five interventional procedures, including high-frequency electrotome, argon plasma coagulation (APC), cryotherapy, stent placement and high-pressure balloon dilatation, were used in this study. Among them, two or more procedures were combined to manage complicated airway stenosis according to stenosis causes, types, position, degree and duration, or functions of distal lung tissue and airway. Forty-two cases were treated with high-frequency electrotome and APC, 54 cases with high-frequency electrotome, cryotherapy and APC, 29 cases with high-frequency electrotome, cryotherapy, APC and stent placement, and 13 cases with cryotherapy, APC and high-pressure balloon dilatation. Airways opening, short-term therapeutic effects and improvement of pulmonary functions were evaluated when ideal curative effects were achieved in the first month after intervention. RESULTS: the total short-term effective rate in the 138 patients was 100%. The airway diameter was increased from (2.6 ± 1.5) mm before operation to (6.2 ± 1.7) mm after operation (P < 0.05). Dyspnea score was decreased from (2.4 ± 0.8) before operation to (0.7 ± 0.6) after operation (P < 0.05). FEV(1) was increased from (1.8 ± 0.6) L before operation to (3.1 ± 0.7) L after operation (P < 0.05). Among 23 cases benign disease, including 4 benign tumor, 15 tuberculosis and 4 other granulomatosis, 5 cases with various degrees of restenosis needed further interventional therapy after 3 months of follow up and effective rate was 78.3% (18/23). After 6 months of follow-up, 3 cases with restenosis needed re-intervention, and the effective rate was 86.9% (20/23). All of the 23 cases did not experience stenosis after 12 months of follow-up. Patients with malignant tumor were not followed up for long term. CONCLUSION: combination of five interventional procedures, including high-frequency electrotome, APC, cryotherapy, stent placement and high-pressure balloon dilatation, has fewer complications and favorable clinical effects in management of intractable central airway stenosis.

Detection of lung adenocarcinoma using magnetic beads based matrix-assisted laser desorption/ionization time-of-flight mass spectrometry serum protein profiling.

BACKGROUND: Recently, due to the rapid development of proteomic techniques, great advance has been made in many scientific fields. We aimed to use magnetic beads (liquid chip) based matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF-MS) technology to screen distinctive biomarkers for lung adenocarcinoma (adCA), and to establish the diagnostic protein profiles. METHODS: Using weak cation exchange magnetic beads (MB-WCX) to isolate and purify low molecular weight proteins from sera of 35 lung adCA, 46 benign lung diseases (BLDs) and 44 healthy individuals. The resulting spectra gained by anchor chip-MALDI-TOF-MS were analyzed by ClinProTools and a pattern recognition genetic algorithm (GA). RESULTS: In the working mass range of 800 - 10 000 Da, 99 distinctive peaks were resolved in lung adCA versus BLDs, while 101 peaks were resolved in lung adCA versus healthy persons. The profile gained by GA that could distinguish adCA from BLDs was comprised of 4053.88, 4209.57 and 3883.33 Da with sensitivity of 80%, specificity of 93%, while that could separate adCA from healthy control was comprised of 2951.83 Da and 4209.73 Da with sensitivity of 94%, specificity of 95%. The sensitivity provided by carcinoembryonic antigen (CEA) in this experiment was significantly lower than our discriminatory profiles (P < 0.005). We further identified a eukaryotic peptide chain release factor GTP-binding subunit (eRF3b) (4209 Da) and a complement C3f (1865 Da) that may serve as candidate biomarkers for lung adCA. CONCLUSION: Magnetic beads based MALDI-TOF-MS technology can rapidly and effectively screen distinctive proteins/polypeptides from sera of lung adCA patients and controls, which has potential value for establishing a new diagnostic method for lung adCA.

[Application of liquid chip-mass spectrometry technology for screening serum biomarker proteins in lung squamous cell carcinoma].

OBJECTIVE: To screen the serum biomarker proteins of lung squamous cell carcinoma (SCCs) by liquid chip-mass spectrometry technology. METHODS: All serum samples, including 34 SCCs, 46 benign lung diseases (BLDs) and 44 healthy individuals, were analyzed by CLINPROT system in order to study the serum protein expression profiles. Then the discriminatory proteins were detected by FlexAnalysis 3.0 software. Biomarkers were identified by liquid chromatography-tandem mass spectrometry (LCMS/MS). RESULTS: Comparing the differential serum expression proteins between SCCs and healthy individuals, and SCCs and BLDs respectively. Ninety-six differential protein peaks [mass-to-charge ration (M/Z) between 800 and 10 000] were found between SCCs and healthy individuals. In these protein peaks, the expression of protein peaks at 4054.13 M/Z and 4267.46 M/Z had the largest difference between them. The two protein peaks could accurately separate SCCs from healthy individuals by the frame of axes. Similarly, 99 differential protein peaks were automatically detected between SCCs and BLDs. In these protein peaks, the expression of protein peaks at 5065.27 M/Z and 4054.02 M/Z had the largest difference between them. The two protein peaks could accurately separate SCCs from BLDs by the frame of axes. Identified by LC-MS/MS, 1778 M/Z and 1865 M/Z might be assayed jointly and corresponded to complements C3 fragment or C3f precursor. CONCLUSIONS: Differential protein expressions existed between SCCs versus healthy individuals and SCCs versus BLD patients. It is feasible to screen the diagnostic serum biomarkers of SCC with a high sensitivity and specificity by using CLINPROT system.

[Application of laser capture microdissection and surface-enhanced laser desorption ionization time-of-flight mass spectrometry to screen biomarkers for early diagnosis of lung adenocarcinoma].

OBJECTIVE: To improve diagnostic methods to screen biomarkers for early diagnosis in lung adenocarcinoma by employing laser capture microdissection (LCM) and surface-enhanced laser desorption ionization time-of-flight mass spectrometry (SELDI-TOF-MS) and support vector machine (SVM). METHODS: Frozen sections of thickness 8 microm were made using 6 cases of fresh lung cancer tissues and 4 cases of matched normal lung tissues. The sections were stained by improved HE solution. The homogeneous adenocarcinoma cells and normal cells were collected by LCM in each sample, and then SELDI profiles based on PBS II+SELDI-TOF-MS (IMAC protein chip) were analyzed using SVM. RESULTS: High quality cell samples were obtained by LCM quickly and precisely from normal specimens and diseased tissues without interstitium, inflammation and necrosis. Eighty four differential protein peaks were found. Top ten of them were identified as candidate biomarkers; six proteins were significantly weakly expressed in lung cancer tissue compared to normal tissues, but the other four protein were over-expressed (P < 0.05). Every candidate biomarker has undergone the blind-cross-test. Each of them can separate the lung cancer from normal samples with a sensitivity of 100% and a specificity of 100%. The 3191 m/z was considered as disease marker of lung adenocarcinoma. CONCLUSION: The method combined LCM with SELDI-TOF-MS may be able to screen potential biomarkers to distinguish lung cancer from healthy tissue with high sensitivity and specificity, which could improve early diagnosis for lung cancer.

[The value of (18)F-FDG PET/CT in differential diagnosis of benign and malignant pulmonary lesions: a Meta-analysis.].

OBJECTIVE: Using Meta analysis to evaluate the value of (18)F-FDG PET/CT ((18)fluorine-fluorodeoxyglucose Positron emission tomography/computed tomography) in differentiating between benign and malignant pulmonary lesions. METHODS: Relevant documentations from PubMed and other 5 databases from 1980 to 2008 were searched, and the eligible literatures according to the inclusive criteria were selected. The statistical information and quality of science were assessed and classified. The data were analyzed using Meta-Disc1.4 software. The diagnostic value of PET/CT in distinguishing benign from malignant pulmonary lesions was evaluated by the pooled sensitivity, specificity, the likelihood ratio (LR) and summary receiver operating characteristic curve (SROC curve) statistical indicators. RESULTS: Seven literatures were collected including 5 in English and 2 in Chinese, and 795 cases were included in the study. Heterogeneity test showed that the homogeneity of the study was good. By using deterministic models to analyze the data, the value of the weighted sensitivity was 95% (93% - 97%), the specificity was 77% (71% - 82%), the positive likelihood ratio was 4.12, negative likelihood ratio was 0.08, and the SROC area under the curve (area under curve, AUC) was 94%. CONCLUSION: PET/CT is of high diagnostic value in differentiation between benign and malignant lung lesions, but large sized, multicenter, prospective studies are needed to assess its clinical value more accurately.

[Use of laser capture microdissection and surface-enhanced laser desorption ionization time-of-flight mass spectrometry to screen differential proteins in lung adenocarcinoma and lung squamous carcinoma].

OBJECTIVE: To screen biomarkers for classification in lung adenocarcinoma and lung squamous carcinoma by using laser capture microdissection (LCM) and surface-enhanced laser desorption ionization time-of-flight mass spectrometry (SELDI-TOF-MS) and support vector machine (SVM). METHODS: Six specimens of lung adenocarcinoma tissues and seven specimens of lung squamous carcinoma tissues obtained during operation were made into frozen sections and stained by improved H-E solution. About 1.2 x 10(5) of homogeneous adenocarcinoma cells and 1.4 x 10(5) of homogeneous lung squamous carcinoma cells were collected using LCM. Then SELDI profiles based on PBS II(+)SELDI-TOF-MS (IMAC protein chip) and the data were analyzed by support vector machine (SVM). RESULTS: Eighty seven differential protein peaks were found and top ten of them were identified as candidate biomarkers. The expression levels of 6 proteins among them with the molecular weights of 3333, 3592, 3848, 5036, 5191, and 5211 respectively in the lung squamous cancer tissues were weaker than those in the adenocarcinoma tissues, and the expression levels of 4 proteins with the molecular weights of 2505, 4004, 4847, and 11 412 in the lung squamous carcinoma tissues were stronger than those in the adenocarcinoma tissues. The expression of the protein with the molecular weight of 4847 in the squamous cancer was significantly stronger than that in the adenocarcinoma (p + 0.032). A discriminatory pattern consisting of 3 proteins with the molecular weights of 4847, 11 412, and 3592 was established with a sensitivity of 100% and a specificity of 100% respectively in separating adenocarcinoma from squamous carcinoma. CONCLUSION: There is a difference in protein component between adenocarcinoma and squamous carcinoma. LCM combined with SELDI-TOF-MS help screen a biomarker pattern to distinguish lung adenocarcinoma from lung squamous carcinoma with high sensitivity and specificity.