scholarly journals Lung Segmentation and Characterization in COVID-19 Patients for Assessing Pulmonary Thromboembolism: An Approach Based on Deep Learning and Radiomics

Electronics ◽  
2021 ◽  
Vol 10 (20) ◽  
pp. 2475
Author(s):  
Vitoantonio Bevilacqua ◽  
Nicola Altini ◽  
Berardino Prencipe ◽  
Antonio Brunetti ◽  
Laura Villani ◽  
...  

The COVID-19 pandemic is inevitably changing the world in a dramatic way, and the role of computed tomography (CT) scans can be pivotal for the prognosis of COVID-19 patients. Since the start of the pandemic, great care has been given to the relationship between interstitial pneumonia caused by the infection and the onset of thromboembolic phenomena. In this preliminary study, we collected n = 20 CT scans from the Polyclinic of Bari, all from patients positive with COVID-19, nine of which developed pulmonary thromboembolism (PTE). For eight CT scans, we obtained masks of the lesions caused by the infection, annotated by expert radiologists; whereas for the other four CT scans, we obtained masks of the lungs (including both healthy parenchyma and lesions). We developed a deep learning-based segmentation model that utilizes convolutional neural networks (CNNs) in order to accurately segment the lung and lesions. By considering the images from publicly available datasets, we also realized a training set composed of 32 CT scans and a validation set of 10 CT scans. The results obtained from the segmentation task are promising, allowing to reach a Dice coefficient higher than 97%, posing the basis for analysis concerning the assessment of PTE onset. We characterized the segmented region in order to individuate radiomic features that can be useful for the prognosis of PTE. Out of 919 extracted radiomic features, we found that 109 present different distributions according to the Mann–Whitney U test with corrected p-values less than 0.01. Lastly, nine uncorrelated features were retained that can be exploited to realize a prognostic signature.

1999 ◽  
Vol 14 (3) ◽  
pp. 163-166 ◽  
Author(s):  
G Bersani ◽  
A Garavini ◽  
I Taddei ◽  
G Tanfani ◽  
M Nordio ◽  
...  

SummaryComputed tomography studies concerning pineal calcification (PC) in schizophrenia have been conducted mainly by one author who correlated this calcification with several aspects of the illness. On the basis of these findings the aim of the present study was to analyze size and incidence of pineal gland calcification by CT in schizophrenics and healthy controls, and to verify the relationship between pineal calcification and age, and the possible correlation with psychopathologic variables. Pineal calcification was measured on CT scans of 87 schizophrenics and 46 controls divided into seven age subgroups of five years each. No significant differences in PC incidence and mean size between patients and controls were observed as far as the entire group was considered. PC size correlated with age both in schizophrenics and controls. We found a higher incidence of PC in schizophrenics in the age subgroup of 21–25 years, and a negative correlation with positive symptoms of schizophrenia in the overall group. These findings could suggest a premature calcific process in schizophrenics and a probable association with `non-paranoid' aspects of the illness. Nevertheless the potential role of this process possibly related to some aspects of the altered neurodevelopment in schizophrenia is still unclear.


2021 ◽  
Vol 2021 ◽  
pp. 1-14
Author(s):  
Kaiming Zhang ◽  
Liqin Ping ◽  
Tian Du ◽  
Yan Wang ◽  
Ya Sun ◽  
...  

Background. Breast cancer was associated with imbalance between oxidation and antioxidation. Local oxidative stress in tumors is closely related to the occurrence and development of breast cancer. However, the relationship between systematic oxidative stress and breast cancer remains unclear. This study is aimed at exploring the prognostic value of systematic oxidative stress in patients with operable breast cancer. Methods. A total of 1583 operable female breast cancer patients were randomly assigned into the training set and validation set. The relationship between systematic oxidative stress biomarkers and prognosis were analyzed in the training and validation sets. Results. The systematic oxidative stress score (SOS) was established based on five systematic oxidative stress biomarkers including serum creatinine (CRE), serum albumin (ALB), total bilirubin (TBIL), lactate dehydrogenase (LDH), and blood urea nitrogen (BUN). SOS was an independent prognostic factor for operable breast cancer patients. A nomogram based on SOS and clinical characteristics could accurately predict the prognosis of operable breast cancer patients, and the area under the curve (AUC) of the nomogram was 0.823 in the training set and 0.872 in the validation set, which was much higher than the traditional prognostic indicators. Conclusions. SOS is an independent prognostic indicator for operable breast cancer patients. A prediction model based on SOS could accurately predict the outcome of operable breast cancer patients.


2016 ◽  
Vol 24 (3) ◽  
pp. 111-116
Author(s):  
Shrikrishna B H ◽  
Jyothi A C

Introduction There are several studies with contradictory findings about the role of concha bullosa with predisposition to rhinosinusitis. This study was conducted to assess the relationship of osteomeatal unit blockage with concha bullosa. Materials and method A cross-sectional observational study by radiological assessment of prevalence of chronic rhinosinusitis and blockage of ipsilateral osteomeatal unit was done on 100 cases of concha bullosa detected on computed tomography to determine the prevalence of chronic rhinosinusitis in subjects with concha bullosa and to examine the latter’s relationship to osteomeatal unit blockage, which is a precursor for rhinosinusitis. Result One hundred cases of concha bullosa were studied in a total of 87 CT scan films depicting concha bullosa. Some CT scans showed unilateral concha bullosa while few scans showed bilateral concha bullosa. Ipsilateral rhinosinusitis was found in only 31% of the sides in scans of subjects with concha bullosa. Of the total 100 concha bullosae studied, extensive type was the commonest followed by bulbous and lamellar variety. Discussion Although rhinosinusitis was more predominant in the extensive type of concha bullosa compared to other types, it was statistically not significant and there was no statistically significant association between any type of concha bullosa with ipsilateral rhinosinusitis either in right side or left side. Conclusion This study has found no statistically significant association between any type of concha bullosa with rhinosinusitis. A bigger study with larger sample size is required to better assess the strength of association between concha bullosa and rhinosinusitis.


2021 ◽  
Vol 23 (1) ◽  
pp. 116-139
Author(s):  
Xuan Li ◽  
Feng Wang

Abstract Although it is widely acknowledged that different speech processes may interact with each other, the way that nasalization affects phonation remains poorly understood. This paper explores the relationship between nasalization and phonation, by analyzing the phonetic cues of the tense/lax distinction both in nasalization and non-nasalization in the Bai language. The data for discussion is from two Bai dialects, Chengbei and Jinhua, which have a tense/lax distinction in both nasalized and non-nasalized syllables. Three phonation parameters – fundamental frequency (F0), open quotient (OQ), and speed quotient (SQ) – are extracted from EGG signals for analysis. It is found that the influence of nasalization on phonation varies with the tone contours. As for the level tones, the role of phonation manner in tone distinction is not evident in nasalization in that tense tones can be distinguished from lax tones only by pitch. However, in non-nasalization, phonation manner plays an indispensable role in tone distinction, in that the contrast between tense and lax tones are reflected not only on F0 but also on OQ and SQ. Moreover, non-nasalized tense tones are more likely to be accompanied by non-modal phonation that is characterized by a significantly higher SQ. In terms of articulatory explanation, high SQ in non-modal phonation is the result of the vibration of tightened vocal folds, and the tension of vocal folds is caused by raising the soft palate in non-nasalization. As for the falling tones, the role of phonation manner in tone distinction is more salient in nasalization than in non-nasalization in Chengbei Bai, but it is not attested in Jinhua Bai. This study shows that the interaction between nasalization and phonation in Bai can be revealed in the analysis of phonation parameters, i.e. F0, OQ, and SQ.


Author(s):  
Nhan T. Nguyen ◽  
Dat Q. Tran ◽  
Nghia T. Nguyen ◽  
Ha Q. Nguyen

AbstractWe propose a novel method that combines a convolutional neural network (CNN) with a long short-term memory (LSTM) mechanism for accurate prediction of intracranial hemorrhage on computed tomography (CT) scans. The CNN plays the role of a slice-wise feature extractor while the LSTM is responsible for linking the features across slices. The whole architecture is trained end-to-end with input being an RGB-like image formed by stacking 3 different viewing windows of a single slice. We validate the method on the recent RSNA Intracranial Hemorrhage Detection challenge and on the CQ500 dataset. For the RSNA challenge, our best single model achieves a weighted log loss of 0.0522 on the leaderboard, which is comparable to the top 3% performances, almost all of which make use of ensemble learning. Importantly, our method generalizes very well: the model trained on the RSNA dataset significantly outperforms the 2D model, which does not take into account the relationship between slices, on CQ500. Our codes and models will be made public.


2018 ◽  
pp. 1-8 ◽  
Author(s):  
Okyaz Eminaga ◽  
Nurettin Eminaga ◽  
Axel Semjonow ◽  
Bernhard Breil

Purpose The recognition of cystoscopic findings remains challenging for young colleagues and depends on the examiner’s skills. Computer-aided diagnosis tools using feature extraction and deep learning show promise as instruments to perform diagnostic classification. Materials and Methods Our study considered 479 patient cases that represented 44 urologic findings. Image color was linearly normalized and was equalized by applying contrast-limited adaptive histogram equalization. Because these findings can be viewed via cystoscopy from every possible angle and side, we ultimately generated images rotated in 10-degree grades and flipped them vertically or horizontally, which resulted in 18,681 images. After image preprocessing, we developed deep convolutional neural network (CNN) models (ResNet50, VGG-19, VGG-16, InceptionV3, and Xception) and evaluated these models using F1 scores. Furthermore, we proposed two CNN concepts: 90%-previous-layer filter size and harmonic-series filter size. A training set (60%), a validation set (10%), and a test set (30%) were randomly generated from the study data set. All models were trained on the training set, validated on the validation set, and evaluated on the test set. Results The Xception-based model achieved the highest F1 score (99.52%), followed by models that were based on ResNet50 (99.48%) and the harmonic-series concept (99.45%). All images with cancer lesions were correctly determined by these models. When the focus was on the images misclassified by the model with the best performance, 7.86% of images that showed bladder stones with indwelling catheter and 1.43% of images that showed bladder diverticulum were falsely classified. Conclusion The results of this study show the potential of deep learning for the diagnostic classification of cystoscopic images. Future work will focus on integration of artificial intelligence–aided cystoscopy into clinical routines and possibly expansion to other clinical endoscopy applications.


2020 ◽  
Vol 10 (2) ◽  
pp. 73 ◽  
Author(s):  
Alex A. Nguyen ◽  
Pedro D. Maia ◽  
Xiao Gao ◽  
Pablo F. Damasceno ◽  
Ashish Raj

Background: The release of a broad, longitudinal anatomical dataset by the Parkinson’s Progression Markers Initiative promoted a surge of machine-learning studies aimed at predicting disease onset and progression. However, the excessive number of features used in these models often conceals their relationship to the Parkinsonian symptomatology. Objectives: The aim of this study is two-fold: (i) to predict future motor and cognitive impairments up to four years from brain features acquired at baseline; and (ii) to interpret the role of pivotal brain regions responsible for different symptoms from a neurological viewpoint. Methods: We test several deep-learning neural network configurations, and report our best results obtained with an autoencoder deep-learning model, run on a 5-fold cross-validation set. Comparison with Existing Methods: Our approach improves upon results from standard regression and others. It also includes neuroimaging biomarkers as features. Results: The relative contributions of pivotal brain regions to each impairment change over time, suggesting a dynamical reordering of culprits as the disease progresses. Specifically, the Putamen is initially the most critical region accounting for the overall cognitive state, only being surpassed by the Substantia Nigra in later years. The Pallidum is the first region to influence motor scores, followed by the parahippocampal and ambient gyri, and the anterior orbital gyrus. Conclusions: While the causal link between regional brain atrophy and Parkinson symptomatology is poorly understood, our methods demonstrate that the contributions of pivotal regions to cognitive and motor impairments are more dynamical than generally appreciated.


Sensors ◽  
2021 ◽  
Vol 21 (17) ◽  
pp. 5878 ◽  
Author(s):  
Fares Bougourzi ◽  
Riccardo Contino ◽  
Cosimo Distante ◽  
Abdelmalik Taleb-Ahmed

Since the appearance of the COVID-19 pandemic (at the end of 2019, Wuhan, China), the recognition of COVID-19 with medical imaging has become an active research topic for the machine learning and computer vision community. This paper is based on the results obtained from the 2021 COVID-19 SPGC challenge, which aims to classify volumetric CT scans into normal, COVID-19, or community-acquired pneumonia (Cap) classes. To this end, we proposed a deep-learning-based approach (CNR-IEMN) that consists of two main stages. In the first stage, we trained four deep learning architectures with a multi-tasks strategy for slice-level classification. In the second stage, we used the previously trained models with an XG-boost classifier to classify the whole CT scan into normal, COVID-19, or Cap classes. Our approach achieved a good result on the validation set, with an overall accuracy of 87.75% and 96.36%, 52.63%, and 95.83% sensitivities for COVID-19, Cap, and normal, respectively. On the other hand, our approach achieved fifth place on the three test datasets of SPGC in the COVID-19 challenge, where our approach achieved the best result for COVID-19 sensitivity. In addition, our approach achieved second place on two of the three testing sets.


Sign in / Sign up

Export Citation Format

Share Document