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
Interest
Topics
Details

Details

  • Name

    Lio Gonçalves
  • Role

    External Research Collaborator
  • Since

    01st January 2014
  • Nationality

    Portugal
  • Contacts

    +351222094106
    lio.goncalves@inesctec.pt
Publications

2022

Semantic Segmentation of Dog's Femur and Acetabulum Bones with Deep Transfer Learning in X-Ray Images

Authors
da Silva, DEM; Filipe, V; Franco-Goncalo, P; Colaco, B; Alves-Pimenta, S; Ginja, M; Goncalves, L;

Publication
INTELLIGENT SYSTEMS DESIGN AND APPLICATIONS, ISDA 2021

Abstract
Hip dysplasia is a genetic disease that causes the laxity of the hip joint and is one of the most common skeletal diseases found in dogs. Diagnosis is performed through an X-ray analysis by a specialist and the only way to reduce the incidence of this condition is through selective breeding. Thus, there is a need for an automated tool that can assist the specialist in diagnosis. In this article, our objective is to develop models that allow segmentation of the femur and acetabulum, serving as a foundation for future solutions for the automated detection of hip dysplasia. The studied models present state-of-the-art results, reaching dice scores of 0.98 for the femur and 0.93 for the acetabulum.

2022

Acetabular Coverage Area Occupied by the Femoral Head as an Indicator of Hip Congruency

Authors
Franco Goncalo, P; da Silva, DM; Leite, P; Alves Pimenta, S; Colaco, B; Ferreira, M; Goncalves, L; Filipe, V; McEvoy, F; Ginja, M;

Publication
ANIMALS

Abstract
Simple Summary Radiographic diagnosis is essential for the genetic control of canine hip dysplasia (HD). The Federation Cynologique Internationale (FCI) scoring HD scheme is based on objective and qualitative radiographic criteria. Subjective interpretations can lead to errors in diagnosis and, consequently, to incorrect selective breeding, which in turn impacts the gene pool of dog breeds. The aim of this study was to use a computer method to calculate the Hip Congruency Index (HCI) to objectively estimate radiographic hip congruency for future application in the development of computer vision models capable of classifying canine HD. The HCI measures the percentage of acetabular coverage that is occupied by the femoral head. Normal hips are associated with an even, parallel joint surface that translates into reduced acetabular free space, which increases with hip subluxation and becomes maximal in hip dislocation. We found statistically significant differences in mean HCI values among all five FCI categories. These results demonstrate that the HCI reliably reflects the different degrees of congruency associated with HD. Therefore, it is expected that when used in conjunction with other HD evaluation parameters, such as Norberg angle and assessment of osteoarthritic signs, it can improve the diagnosis by making it more accurate and unequivocal. Accurate radiographic screening evaluation is essential in the genetic control of canine HD, however, the qualitative assessment of hip congruency introduces some subjectivity, leading to excessive variability in scoring. The main objective of this work was to validate a method-Hip Congruency Index (HCI)-capable of objectively measuring the relationship between the acetabulum and the femoral head and associating it with the level of congruency proposed by the Federation Cynologique Internationale (FCI), with the aim of incorporating it into a computer vision model that classifies HD autonomously. A total of 200 dogs (400 hips) were randomly selected for the study. All radiographs were scored in five categories by an experienced examiner according to FCI criteria. Two examiners performed HCI measurements on 25 hip radiographs to study intra- and inter-examiner reliability and agreement. Additionally, each examiner measured HCI on their half of the study sample (100 dogs), and the results were compared between FCI categories. The paired t-test and the intraclass correlation coefficient (ICC) showed no evidence of a systematic bias, and there was excellent reliability between the measurements of the two examiners and examiners' sessions. Hips that were assigned an FCI grade of A (n = 120), B (n = 157), C (n = 68), D (n = 38) and E (n = 17) had a mean HCI of 0.739 +/- 0.044, 0.666 +/- 0.052, 0.605 +/- 0.055, 0.494 +/- 0.070 and 0.374 +/- 0.122, respectively (ANOVA, p < 0.01). Therefore, these results show that HCI is a parameter capable of estimating hip congruency and has the potential to enrich conventional HD scoring criteria if incorporated into an artificial intelligence algorithm competent in diagnosing HD.

2022

Early hip laxity screening and later canine hip dysplasia development

Authors
Santana, A; Alves-Pimenta, S; Franco-Goncalo, P; Goncalves, L; Martins, J; Colaco, B; Ginja, M;

Publication
VETERINARY WORLD

Abstract
Background and Aim: Passive hip laxity (PHL) is considered the primary risk factor for canine hip dysplasia (HD) and is estimated, in stress hip radiographs, using the distraction index (DI). The study aimed to associate the early PHL using the hip Distractor of University of Tras-os-Montes and Alto Douro (DisUTAD) and the late HD grades. Materials and Methods: A total of 41 dogs (82 hips) were submitted to a follow-up study. First, between 4 and 12 months of age, dogs were radiographed using the DisUTAD hip distractor and were determined the DI for each hip joint. Then, after 12 months of age, dogs were reevaluated for HD using the conventional hip ventrodorsal projection and hips were evaluated for HD using the Federation Cynologique Internationale (FCI) scoring system. Results: Hips of dogs' in the second examination with FCI grades of A (n=28), B (n=11), C (n=22), and D and E (n=21) had an early DI of 0.32 +/- 0.1, 0.380.08, 0.50 +/- 0.12, and 0.64 +/- 0.11, respectively. Statistical analysis using the general linear model univariate, with the DI as dependent variable and the FCI grades, side and sex as fixed factors, and the post hoc Bonferroni correction test showed significant differences among FCI grades (p<0.05). Conclusion: These results show the association between early DI and the late FCI HD grades and the DisUTAD is recommended for the early canine HD diagnosis.

2021

Refinement of animal model of colorectal carcinogenesis through the definition of novel humane endpoints

Authors
Silva Reis, R; Faustino Rocha, AI; Goncalves, M; Ribeiro, CC; Ferreira, T; Ribeiro Silva, C; Goncalves, L; Antunes, L; Venancio, C; Ferreira, R; Gama, A; Oliveira, PA;

Publication
ANIMALS

Abstract
This study aimed to define appropriate humane endpoints (HEs) for an animal model of colorectal carcinogenesis (CRC). Twenty-nine male Wistar rats were divided into two control groups (CTRL1 and CTRL2) injected with ethylenediamine tetraacetic acid (EDTA)–saline solutions and two induced groups (CRC1 and CRC2) injected with 1,2-dimethylhydrazine (DMH) for seven weeks. A score sheet with 14 biological parameters was used to assess animal welfare. Groups CRC1 and CTRL1 and groups CRC2 and CTRL2 were euthanized 11 and 17 weeks after the first DMH administration, respectively. Five animals from the induced groups died unexpectedly during the protocol (survival rates of 75.0% and 66.7% for groups CRC1 and CRC2, respectively). The final mean body weight (BW) was smaller in the CRC groups when compared with that in the CTRL groups. A uniformity of characteristics preceding the premature animals’ death was observed, namely an increase of 10% in mean BW, swollen abdomen, diarrhea, and priapism. The surface abdominal temperature of group CRC2 was significantly higher, when compared with that of group CTRL2. The parameters already described in other cancer models proved to be insufficient. For the CRC model, we considered assessing the abdominal temperature, priapism, and sudden increase in the BW. © 2021 by the authors. Licensee MDPI, Basel, Switzerland.

2020

Distraction index measurement on the dog's hip joint using a dedicated software

Authors
Alves Pimenta, S; Santana, A; Martins, J; Colaco, B; Goncalves, L; Ginja, M;

Publication
ARQUIVO BRASILEIRO DE MEDICINA VETERINARIA E ZOOTECNIA

Abstract
The aim of this study was to test the accuracy of a new automated computer software tool for the assessment of passive hip laxity. The hip laxity was estimated using the dedicated computer software by two blinded evaluators, one previously trained and one without specific training for distraction index measurement, in two independent sessions using 230 hip joints from 115 dogs that underwent screening for passive hip laxity using the distraction view. Previously, all of these radiographs were sent to PennHIP Analysis Center for an official distraction index record. The measurement repeatability of the two sessions was adequate for both evaluators. The reproducibility of the official distraction index measurement, mean distraction index +/- standard deviation 0.44 +/- 0.15, was adequate (P>0.05) for the trained evaluator, 0.44 +/- 0.15, and non-adequate (P<0.05), for the non-trained evaluator 0.47 +/- 0.17. The distraction index measurement tool proposed can be used with confidence for hip laxity evaluation by trained evaluators, as it provided good repeatability and reproducibility of official reports. The simplicity of the process described leads to a less time-consuming and more affordable procedure.

Supervised
thesis

2016

Desenvolvimento de um sistema wearable para a monitorização da taxa de alcoolemia

Author
Ana Patrícia Queirós Gomes

Institution
UTAD

2016

Monitorização não invasiva do nível de lactato

Author
Sónia Isabel Martins Pereira

Institution
UTAD

2015

Desenvolvimento da aplicação informática interpharm para estudar a interção entre fármacos

Author
Ana Rita Fernandes de Sousa

Institution
UTAD