Fighting Together against the Pandemic: Learning Multiple Models on Tomography Images for COVID-19 Diagnosis

Manzo, Mario; Pellino, Simone

doi:10.3390/ai2020016

Cited by 18 publications

(9 citation statements)

References 22 publications

(33 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To address the above issues, transfer learning was adopted to allow the DeepLabV3+ to be able to segment grapes with high accuracy on the condition that only a small dataset was provided. The basic idea of transfer learning is to transfer the knowledge gained in other fields to a new task [25]. In order to implement the transfer learning, the DeepLabV3+ was first trained on the PASCAL VOC2012, which contained tens of thousands of images with 20 classes.…”

Section: Transfer Learningmentioning

confidence: 99%

Segmentation of Overlapping Grape Clusters Based on the Depth Region Growing Method

2021

View full text Add to dashboard Cite

Accurately extracting the grape cluster at the front of overlapping grape clusters is the primary problem of the grape-harvesting robot. To solve the difficult problem of identifying and segmenting the overlapping grape clusters in the cultivation environment of a trellis, a simple method based on the deep learning network and the idea of region growing is proposed. Firstly, the region of grape in an RGB image was obtained by the finely trained DeepLabV3+ model. The idea of transfer learning was adopted when training the network with a limited number of training sets. Then, the corresponding region of the grape in the depth image captured by RealSense D435 was processed by the proposed depth region growing algorithm (DRG) to extract the front cluster. The depth region growing method uses the depth value instead of gray value to achieve clustering. Finally, it fils the holes in the clustered region of interest, extracts the contours, and maps the obtained contours to the RGB image. The images captured by RealSense D435 in a natural trellis environment were adopted to evaluate the performance of the proposed method. The experimental results showed that the recall and precision of the proposed method were 89.2% and 87.5%, respectively. The demonstrated performance indicated that the proposed method could satisfy the requirements of practical application for robotic grape harvesting.

show abstract

Section: Transfer Learningmentioning

confidence: 99%

Segmentation of Overlapping Grape Clusters Based on the Depth Region Growing Method

2021

View full text Add to dashboard Cite

show abstract

“…The neural network ensemble is being used in various fields such as image classification and detection to improve performance and make a robust model [26][27][28][29]. Chen et al [26] classified a channel randomly selected for hyperspectral images and then ensembled each classifier with the majority voting to derive the final result.…”

Section: Rvm and Cnnmentioning

confidence: 99%

“…This model showed a state-of-the-art performance even using a 10× smaller dataset than previous studies. Manzo and Pellino [28] proposed a neural network ensemble model to diagnose COVID-19 disease from computer tomography images. They demonstrated that combining multiple models produces better results than using a single model.…”

Section: Rvm and Cnnmentioning

confidence: 99%

Dual Image-Based CNN Ensemble Model for Waste Classification in Reverse Vending Machine

2021

View full text Add to dashboard Cite

A reverse vending machine motivates citizens to bring recyclable waste by rewarding them, which is a viable solution to increase the recycling rate. Reverse vending machines generally use near-infrared sensors, barcode sensors, or cameras to classify recycling resources. However, sensor-based reverse vending machines suffer from a high configuration cost and the limited scope of target objects, and conventional single image-based reverse vending machines usually make erroneous predictions about intentional fraud objects. This paper proposes a dual image-based convolutional neural network ensemble model to address these problems. For this purpose, we first created a prototype reverse vending machine and constructed an image dataset containing two cross-sections of objects, top and front view. Then, we chose convolutional neural network models widely used in image classification as the candidates for building an accurate and lightweight ensemble model. Considering the size and classification performance of candidates, we constructed the best-fit ensemble combination and evaluated its classification performance. The final ensemble model showed a classification accuracy higher than 95% for all target classes, including fraud objects. This result proves that our approach achieves better robustness against intentional fraud objects than single image-based models and thus can broaden the scope for target resources. The measurement results on lightweight embedded platforms also demonstrated that our model provides a short inference time that is enough to facilitate the real-time execution of reverse vending machines based on low-cost edge artificial intelligence devices.

show abstract

“…We used a standard thorax model from Ercleve [24] as a reference to determine thoracic abnormalities in COVID-19 sufferers. Several other research that use thorax imagery in identifying COVID-19 are the use of the internet of things (IoT) [25], adaptive stratification [26], social approachment [27], blood cell ratio [28], thoraxic surgery [29], tomography images [30], otolaryngology surgery [31], handheld optical system [32], SOM-LWL method [33], therapeutic effects [34], upper respiratory tract [35].…”

Section: Introductionmentioning

confidence: 99%

Morphological characteristics of X-ray thorax images of COVID-19 patients using the Bradley thresholding segmentation

Supriyanti

Alqaaf

Ramadhani

et al. 2021

IJEECS

View full text Add to dashboard Cite

The coronavirus disease 2019 (COVID-19) pandemic has made test screening much needed. Currently, the most commonly used is the swab type. Although in fact, there is also a screening method with chest radiology. The purpose of this study is to develop a COVID-19 early detection system based on X-ray images of the patient's thorax in the form of a computer-aided diagnosis. This case is based on the fact that X-ray modalities are available in several health care centers in Indonesia, compared to other modalities such as computed tomography (CT) scan or magnetic resonance imaging (MRI). In this paper, we emphasize the X-ray thorax image segmentation process to explore the morphological information of the thorax. We use the Bradley thresholding segmentation method. The results obtained are promising to be further developed with a performance percentage of 73.33% for the thorax for COVID-19 patients and 54% for the thorax for normal patients.

show abstract

Fighting Together against the Pandemic: Learning Multiple Models on Tomography Images for COVID-19 Diagnosis

Cited by 18 publications

References 22 publications

Segmentation of Overlapping Grape Clusters Based on the Depth Region Growing Method

Segmentation of Overlapping Grape Clusters Based on the Depth Region Growing Method

Dual Image-Based CNN Ensemble Model for Waste Classification in Reverse Vending Machine

Morphological characteristics of X-ray thorax images of COVID-19 patients using the Bradley thresholding segmentation

Contact Info

Product

Resources

About