Cookies Policy
The website need some cookies and similar means to function. If you permit us, we will use those means to collect data on your visits for aggregated statistics to improve our service. Find out More
Accept Reject
  • Menu
Publications

Publications by CTM

2022

Lung Segmentation in CT Images: A Residual U-Net Approach on a Cross-Cohort Dataset

Authors
Sousa, J; Pereira, T; Silva, F; Silva, MC; Vilares, AT; Cunha, A; Oliveira, HP;

Publication
APPLIED SCIENCES-BASEL

Abstract
Lung cancer is one of the most common causes of cancer-related mortality, and since the majority of cases are diagnosed when the tumor is in an advanced stage, the 5-year survival rate is dismally low. Nevertheless, the chances of survival can increase if the tumor is identified early on, which can be achieved through screening with computed tomography (CT). The clinical evaluation of CT images is a very time-consuming task and computed-aided diagnosis systems can help reduce this burden. The segmentation of the lungs is usually the first step taken in image analysis automatic models of the thorax. However, this task is very challenging since the lungs present high variability in shape and size. Moreover, the co-occurrence of other respiratory comorbidities alongside lung cancer is frequent, and each pathology can present its own scope of CT imaging appearances. This work investigated the development of a deep learning model, whose architecture consists of the combination of two structures, a U-Net and a ResNet34. The proposed model was designed on a cross-cohort dataset and it achieved a mean dice similarity coefficient (DSC) higher than 0.93 for the 4 different cohorts tested. The segmentation masks were qualitatively evaluated by two experienced radiologists to identify the main limitations of the developed model, despite the good overall performance obtained. The performance per pathology was assessed, and the results confirmed a small degradation for consolidation and pneumocystis pneumonia cases, with a DSC of 0.9015 ± 0.2140 and 0.8750 ± 0.1290, respectively. This work represents a relevant assessment of the lung segmentation model, taking into consideration the pathological cases that can be found in the clinical routine, since a global assessment could not detail the fragilities of the model.

2022

Myope Models - Are face presentation attack detection models short-sighted?

Authors
Neto, PC; Sequeira, AF; Cardoso, JS;

Publication
2022 IEEE/CVF WINTER CONFERENCE ON APPLICATIONS OF COMPUTER VISION WORKSHOPS (WACVW 2022)

Abstract

2022

Editorial of the Special Issue from WorldCIST'20

Authors
Domingues, I; Sequeira, AF;

Publication
COMPUTATIONAL AND MATHEMATICAL ORGANIZATION THEORY

Abstract

2022

Proof of Concept of a Low-Cost Beam-Steering Hybrid Reflectarray that Mixes Microstrip and Lens Elements Using Passive Demonstrators

Authors
Luo, Q; Gao, S; Hu, W; Liu, W; Pessoa, LM; Sobhy, M; Sun, YC;

Publication
IEEE COMMUNICATIONS MAGAZINE

Abstract
In this article, a proof-of-concept study on the use of a hybrid design technique to reduce the number of phase shifters of a beam-scanning reflectarray (RA) is presented. An extended hemispherical lens antenna with feeds inspired by the retrodirective array is developed as a reflecting element, and the hybrid design technique mixes the lenses with the microstrip patch elements to realize a reflecting surface. Compared to the conventional designs that only use microstrip antennas to realize a reflecting surface, given a fixed aperture size the presented design uses 25 percent fewer array elements while shows comparable beam-steering performance. As a result of using fewer elements, the number of required phase shifters or other equivalent components such as RF switches and tunable materials is reduced by 25 percent, which leads to the reduction of the overall antenna system's complexity, cost, and power consumption. To verify the design concept, two passive prototypes with a center frequency at 12.5 GHz were designed and fabricated. The reflecting surface was fabricated by using standard PCB manufacturing and the lenses were fabricated using 3D printing. Good agreement between the simulation and measurement results is obtained. The presented design concept can be extended to the design of RAs operating at different frequency bands including millimetre-wave frequencies with similar radiation performances. The presented design method is not limited to the microstrip patch reflecting elements and can also be applied to the design of the hybrid RAs with different types of reflecting elements.

2022

Learning Models for Traumatic Brain Injury Mortality Prediction on Pediatric Electronic Health Records

Authors
Fonseca, J; Liu, XY; Oliveira, HP; Pereira, T;

Publication
FRONTIERS IN NEUROLOGY

Abstract
BackgroundTraumatic Brain Injury (TBI) is one of the leading causes of injury related mortality in the world, with severe cases reaching mortality rates of 30-40%. It is highly heterogeneous both in causes and consequences, complicating medical interpretation and prognosis. Gathering clinical, demographic, and laboratory data to perform a prognosis requires time and skill in several clinical specialties. Machine learning (ML) methods can take advantage of the data and guide physicians toward a better prognosis and, consequently, better healthcare. The objective of this study was to develop and test a wide range of machine learning models and evaluate their capability of predicting mortality of TBI, at hospital discharge, while assessing the similarity between the predictive value of the data and clinical significance.MethodsThe used dataset is the Hackathon Pediatric Traumatic Brain Injury (HPTBI) dataset, composed of electronic health records containing clinical annotations and demographic data of 300 patients. Four different classification models were tested, either with or without feature selection. For each combination of the classification model and feature selection method, the area under the receiver operator curve (ROC-AUC), balanced accuracy, precision, and recall were calculated.ResultsMethods based on decision trees perform better when using all features (Random Forest, AUC = 0.86 and XGBoost, AUC = 0.91) but other models require prior feature selection to obtain the best results (k-Nearest Neighbors, AUC = 0.90 and Artificial Neural Networks, AUC = 0.84). Additionally, Random Forest and XGBoost allow assessing the feature's importance, which could give insights for future strategies on the clinical routine.ConclusionPredictive capability depends greatly on the combination of model and feature selection methods used but, overall, ML models showed a very good performance in mortality prediction for TBI. The feature importance results indicate that predictive value is not directly related to clinical significance.

2022

Optimizing Packet Reception Rates for Low Duty-Cycle BLE Relay Nodes

Authors
Paulino, N; Pessoa, LM; Branquinho, A; Almeida, R; Ferreira, I;

Publication
IEEE SENSORS JOURNAL

Abstract
In order to achieve the full potential of the Internet-of-Things, connectivity between devices should be ubiquitous and efficient. Wireless mesh networks are a critical component to achieve this ubiquitous connectivity for a wide range of services, and are composed of terminal devices (i.e., nodes), such as sensors of various types, and wall powered gateway devices, which provide further internet connectivity (e.g., via Wi-Fi). When considering large indoor areas, such as hospitals or industrial scenarios, the mesh must cover a large area, which introduces concerns regarding range and the number of gateways needed and respective wall cabling infrastructure, including data and power. Solutions for mesh networks implemented over different wireless protocols exist, like the recent Bluetooth Low Energy (BLE) 5.1. While BLE provides lower power consumption, some wall-power infrastructure may still be required. Alternatively, if some nodes are battery powered, concerns such as lifetime and packet delivery are introduced. We evaluate a scenario where the intermediate nodes of the mesh are battery powered, using a BLE relay of our own design, which acts as a range extender by forwarding packets from end-nodes to gateways. We present the relay's design and experimentally determine the packet forwarding efficiency for several scenarios and configurations. In the best case, up to 35% of the packets transmitted by 11 end-nodes can be forwarded to a gateway by a single relay under continuous operation. A battery lifetime of 1 year can be achieved with a relay duty cycle of 20%.

  • 1
  • 293