Human Action Recognition (HAR) is a challenging task used in sports such as volleyball, basketball, soccer, and tennis to detect players and recognize their actions and teams' activities during ...training, matches, warm-ups, or competitions. HAR aims to detect the person performing the action on an unknown video sequence, determine the action's duration, and identify the action type. The main idea of HAR in sports is to monitor a player's performance, that is, to detect the player, track their movements, recognize the performed action, compare various actions, compare different kinds and skills of acting performances, or make automatic statistical analysis.
As an action that can occur in the sports field refers to a set of physical movements performed by a player in order to complete a task using their body or interacting with objects or other persons, actions can be of different complexity. Because of that, a novel systematization of actions based on complexity and level of performance and interactions is proposed.
The overview of HAR research focuses on various methods performed on publicly available datasets, including actions of everyday activities. That is just a good starting point; however, HAR is increasingly represented in sports and is becoming more directed towards recognizing similar actions of a particular sports domain. Therefore, this paper presents an overview of HAR applications in sports primarily based on Computer Vision as the main contribution, along with popular publicly available datasets for this purpose.
Machine Learning, Human Action Recognition, Action systematization, Sports Dataset, Human Action Recognition in Sports, Sport.
This paper focuses on image and video content analysis of handball scenes and applying deep learning methods for detecting and tracking the players and recognizing their activities. Handball is a ...team sport of two teams played indoors with the ball with well-defined goals and rules. The game is dynamic, with fourteen players moving quickly throughout the field in different directions, changing positions and roles from defensive to offensive, and performing different techniques and actions. Such dynamic team sports present challenging and demanding scenarios for both the object detector and the tracking algorithms and other computer vision tasks, such as action recognition and localization, with much room for improvement of existing algorithms. The aim of the paper is to explore the computer vision-based solutions for recognizing player actions that can be applied in unconstrained handball scenes with no additional sensors and with modest requirements, allowing a broader adoption of computer vision applications in both professional and amateur settings. This paper presents semi-manual creation of custom handball action dataset based on automatic player detection and tracking, and models for handball action recognition and localization using Inflated 3D Networks (I3D). For the task of player and ball detection, different configurations of You Only Look Once (YOLO) and Mask Region-Based Convolutional Neural Network (Mask R-CNN) models fine-tuned on custom handball datasets are compared to original YOLOv7 model to select the best detector that will be used for tracking-by-detection algorithms. For the player tracking, DeepSORT and Bag of tricks for SORT (BoT SORT) algorithms with Mask R-CNN and YOLO detectors were tested and compared. For the task of action recognition, I3D multi-class model and ensemble of binary I3D models are trained with different input frame lengths and frame selection strategies, and the best solution is proposed for handball action recognition. The obtained action recognition models perform well on the test set with nine handball action classes, with average F1 measures of 0.69 and 0.75 for ensemble and multi-class classifiers, respectively. They can be used to index handball videos to facilitate retrieval automatically. Finally, some open issues, challenges in applying deep learning methods in such a dynamic sports environment, and direction for future development will be discussed.
Systemic sclerosis (SSc) is characterized by the presence of SSc-specific or SSc-associated antibodies (SSc-Abs): anti-topoisomerase I (ATA), anti-centromere (ACA), anti-RNA polymerase III (ARA), ...anti-U3RNP (U3RNP), anti-U1RNP (U1RNP), anti-PmScl (PmScl), anti-Ku (Ku) and anti-Th/To (Th/To), each being associated with specific clinical features and prognosis. The detection of more than one SSc-Abs in SSc patients is rare and only few data about these patients' clinical phenotype is available. The aim of our study was to evaluate the frequency and the disease's features associated with the presence of > 1 SSc-Abs positivity in a large cohort of SSc patients. The autoantibody profiles of 2799 SSc patients from February 2001 to June 2017 were retrospectively reviewed. Patients with > 1 SSc-Abs were identified. Clinical features were collected and compared to a large historical cohort of SSc patients with single SSc-Ab positivity. SSc patients were excluded if previously treated with rituximab, intravenous immunoglobulins or stem cell transplantation. Non-parametric tests were used for statistical analysis. Nearly 5% of SSc patients from our cohort had ≥ 2 autoantibody positivity, and 2.3% (n = 72) had ≥ 2 SSc-Abs positivity. Th e most common combination was U1RNP and ATA (35%). These patients were younger than patients with single autoantibody positivity and showed more commonly a diffuse cutaneous SSc form. They also had higher rates of overlap features compared to ATA patients. Other combinations included U1RNP and ACA (13%), ATA and ACA (7%) and U1RNP and PmScl (5%). In our study we observed that, while infrequently, SSc patients can present with a combination of two SSc-Abs and that the double positivity can influence their clinical phenotype compared to patients with single SSc-Ab positivity. The importance of re-testing SSc-Abs in patients with changing clinical phenotypes was also highlighted, as this may confer a differing risk stratification.
Implantable cardioverter-defibrillator (ICD) therapy remains a corner stone of sudden cardiac death (SCD) prevention in patients with arrhythmogenic right ventricular cardiomyopathy (ARVC). We aimed ...to assess predictors of appropriate ICD therapies in the Scandinavian cohort of ARVC patients who received ICD for primary prevention of SCD. Study group comprised of 79 definite ARVC patients by 2010 Task Force criteria (60% male, age at ICD implant 39 ± 14 years) who were enrolled in the Nordic ARVC Registry and received an ICD for primary SCD prevention. The primary end point of appropriate ICD shock or death from any cause was assessed and compared with 137 definite ARVC patients who received ICD for secondary SCD prevention (74% male, age at ICD implant 42 ± 15 years). In the study group, 38% were ≤35 years of age at baseline, 25% had nonsustained ventricular tachycardia, and 29% had syncope at baseline. Major repolarization abnormality (hazard ratio = 4.00, 95% confidence interval 1.30 to 12.30, p = 0.015) and age ≤35 years (hazard ratio = 4.21, 95% confidence interval 1.49 to 11.85, p = 0.001) independently predicted the primary end point. The outcome did not differ between the primary prevention patients with either of these risk factors and the secondary prevention cohort (2% to 4% annual event rate) whereas patients without risk factors did not have any appropriate ICD shocks during follow-up. In conclusion, young age at ARVC diagnosis and major repolarization abnormality independently predict ICD shocks or death in the primary prevention ICD recipients and associated with the event rate similar to the one observed in the secondary prevention cohort. Our data indicate the benefit of ICD for primary prevention in patients with any of these risk factors.
Abstract
Background
Systemic sclerosis (SSc) is typically manifests with SSc-specific antibodies (SSc-Abs): anti-topoisomerase I (ATA), anti-centromere (ACA), anti-RNA polymerase III (ARA), ...anti-U3RNP (u3RNP), anti-U1RNP (U1RNP), anti-PmScl (PmScl), anti-Ku (Ku) and anti-Th/T0 (Th/T0), each being characterised by distinct clinical features and prognosis. The presence of > 1 SSc-Abs is rare with minimum data about these patients’ clinical phenotype in SSc
Methods
The autoantibody profiles of 2,799 SSc patients from February 2001 to June 2017 were retrospectively reviewed. Patients with >1 SSc-Abs were identified. Clinical features were collected and compared to historical cohorts of SSc patients with single SSc-Ab positivity. Patients were excluded if treated prior to their immunology test with rituximab, iv immunoglobulins or stem cell transplantation. Statistical analysis was performed using Fisher exact test.
Results
72 patients (2.6%) with >1 SSc-Ab were identified. Full clinical data were available for 63 patients. 60 patients (2.1%) had double Ab positivity and 3 patients had triple Ab positivity (0.1%). 13 Ab combinations were present. U1RNP and ATA was the most frequent combination (35%), these patients were significantly younger(51.38 years) than both U1RNP (58.64 years,p=0.050) and ATA (62.03 years,p=0.002) patients and more commonly of diffuse subset (dcSSc)(p = 0.001 and p = 0.041 respectively). Compared to ATA patients overlap features were more frequent (43% vs 15%, p = 0.004) including inflammatory arthritis (p = 0.025) and myositis (p = 0.013) (Table 1). U1RNP and ACA had a significantly higher prevalence of pulmonary arterial hypertension compared to U1RNP (p = 0.039) and ACA (p = 0.022) patients, and compared to ACA patients they were younger (57.88 vs 68.75,p=0.015) with a higher frequency of myositis (p = 0.001). U1RNP and ARA patients were more frequently dcSSc subtype compared to U1RNP patients (75% vs 21%, p = 0.040). U1RNP and PmScl patients had a higher prevalence of myositis compared to U1RNP patients (p = 0.006). ATA and ACA patients behaved similarly to ATA patients with a significantly higher prevalence of lung fibrosis (p = 0.006) and myositis (p = 0.041) compared to ACA. ACA and PmScl (7%) had higher prevalence of myositis compared to ACA patients (p = 0.04).
P155 Table 1
Double antibody group
Comparator group
Clinical features
Arthritis
ILD
PAH
SRC
Cardiac
Myositis
n
p
n
p
n
p
n
p
n
p
n
p
U1 RNP and ATA (n = 21)
6 (29%)
15 (68%)
1 (9%)
0
0
4 (19%)
U1RNP
14 (18%)
1
38 (50%)
0.136
12 (16%)
0.287
1 (1%)
1
2 (3%)
1
12 (16%)
0.744
ATA
28 (10%)
0.025
213 (79%)
0.382
12 (4%)
1
17 (6%)
0.622
20 (7%)
0.378
10 (4%)
0.013
U1RNP and ACA (n = 8)
2 (25%)
2 (25%)
4 (50%)
0
0
3 (37%)
U1RNP
14 (18%)
1
38 (50%)
0.267
12 (16%)
0.039
1 (1%)
1
2 (3%)
1
12 (16%)
0.148
ACA
23 (6%)
0.087
43 (11%)
0.233
56 (15%)
0.022
6 (2%)
1
10 (3%)
1
3 (1%)
0.001
U1RNP and ARA (n = 4)
0
2 (50%)
0
1 (25%)
1 (25%)
0
U1RNP
14 (18%)
0.311
38 (50%)
1
12 (16%)
1
1 (1%)
0.098
2 (3%)
0.144
12 (16%)
1
ARA
4 (3%)
1
64 (45%)
1
15 (10%)
1
34 (24%)
1
4 (3%)
0.131
5 (3%)
1
U1 RNP and PmScl (n = 3)
1 (33%)
2 (66%)
0
0
0
3 (100%)
U1RNP
14 (18%)
1
38 (50%)
1
12 (16%)
1
1 (1%)
1
2 (3%)
1
12 (16%)
0.006
PmScl
7 (13%)
1
27 (50%)
1
3 (6%)
1
3 (6%)
1
1 (2%)
1
21 (39%)
0.069
ATA and ACA (n = 4)
0
3 (75%)
0
0
0
1 (25%)
ATA
28 (10%)
1
213 (79%)
1
12 (4%)
1
17 (6%)
1
20 (7%)
1
10 (4%)
0.153
ACA
23 (6%)
1
43 (11%)
0.006
56 (15%)
1
6 (2%)
1
10 (3%)
1
3 (1%)
0.041
ACA and PmScl (n = 4)
1 (25%)
1 (25%)
0
0
0
1 (25%)
ACA
23 (6%)
0.227
43 (11%)
0.385
56 (15%)
1
6 (2%)
1
10 (3%)
1
3 (1%)
0.04
PmScl
7 (13%)
0.457
27 (50%)
0.612
3 (6%)
1
3 (6%)
1
1 (2%)
1
21 (39%)
1
ACA and U3RNP (n = 2)
0
1 (50%)
0
0
0
0
ACA
23 (6%)
1
43 (11%)
0.216
56 (15%)
1
6 (2%)
1
10 (3%)
1
3 (1%)
1
U3RNP
3 (6%)
1
12 (24%)
0.456
13 (26%)
1
5 (10%)
1
6 (12%)
1
9 (18%)
1
Frequency of clinical features in key double ab group combinations, compared to patients with single SSc specific ab. Significant p values (<0.05) highlighted. ILD (interstitial lung disease), PAH (pulmonary arterial hypertension), SRC (SSc renal crisis)
Conclusion
Coexistence of hallmark autoantibodies is exceedingly rare in SSc patients. When combined, both SSc-Abs have the potential to synergistically interact and modify the clinical phenotype.
Disclosures
C. Campochiaro None. K.E.N. Clark None. L. Host None. A. Sari None. C.P. Denton None. V.H. Ong None.
Komparativna prednost poslovnog subjekta u vrijeme današnje ekonomije definirana je intelektualnim kapitalom kojim svaki pojedini poslovni subjekt raspolaže. Osnovni cilj istraživanja je bio ...istražiti i dokazati meðuzavisnost kvalitete upravljačkog kadra u poslovnom subjektu prema postizanju poslovne izvrsnosti. Naglasak se stavlja na razinu stručne spreme te adekvatno polje obrazovanja upravljačkog kadra u državnim i županijskim lučkim upravama. Metodama intervjuiranja i anketiranja istraživao se upravljački kadar - ravnatelji i predsjednici upravnih vijeća u 6 državnih lučkih uprava te 22 županijske lučke uprave. Rezultati pokazuju kako državne lučke uprave naspram županijskih prednjače u relevantnosti stručnih sprema upravljačkog kadra, što je odmah bilo primjetljivo u samom procesu istraživanja, gdje je prilikom same provedbe državni lučki kadar naspram županijskog shvatio važnost samog istraživanja i mogućnost napretka kroz dobivene smjernice. Moguće je zaključiti kako odgovarajućom stručnom spremom koja donosi odgovarajuća znanja i vještine, neovisno o spolu upravljačkog kadra, može se pospješiti uspješnost rada lučkih uprava.