Prospective study of malaria in pregnancy, placental and congenital malaria in Northwest Colombia.

BACKGROUND: Pregnancy Associated Malaria (PAM) include malaria in pregnancy (MiP), placental malaria (PM), and congenital malaria (CM). The evidence available in Colombia on PAM focuses on one of the presentations (MiP, PM or CM), and no study longitudinally analyses the infection from the pregnant woman, passing through the placenta, until culminating in the newborn. This study determined the frequency of MiP, PM, and CM caused by Plasmodium vivax, Plasmodium falciparum, or mixed infections, according to Thick Blood Smear (TBS) and quantitative Polymerase Chain Reaction (qPCR). Identifying associated factors of PAM and clinical-epidemiological outcomes in northwestern Colombia. METHODS: Prospective study of 431 pregnant women, their placenta, and newborns registered in the data bank of the research Group "Salud y Comunidad César Uribe Piedrahíta" which collected information between 2014 and 2020 in endemic municipalities of the departments of Córdoba and Antioquia. The frequency of infection was determined with 95% confidence intervals. Comparisons were made with the Chi-square test, Student t-test, prevalence ratios, and control for confounding variables by log-binomial regression. RESULTS: The frequency of MiP was 22.3% (4.6% using TBS), PM 24.8% (1.4% using TBS), and CM 11.8% (0% using TBS). Using TBS predominated P. vivax. Using qPCR the proportions of P. vivax and P. falciparum were similar for MiP and PM, but P. falciparum predominated in CM. The frequency was higher in nulliparous, and women with previous malaria. The main clinical effects of PAM were anaemia, low birth weight, and abnormal APGAR score. CONCLUSIONS: The magnitude of infections was not detected with TBS because most cases were submicroscopic (TBS-negative, qPCR-positive). This confirmed the importance of improving the molecular detection of cases. PAM continue being underestimated in the country due to that in Colombia the control programme is based on TBS, despite its outcomes on maternal, and congenital health.

Automatic multi-view pose estimation in focused cardiac ultrasound.

Focused cardiac ultrasound (FoCUS) is a valuable point-of-care method for evaluating cardiovascular structures and function, but its scope is limited by equipment and operator's experience, resulting in primarily qualitative 2D exams. This study presents a novel framework to automatically estimate the 3D spatial relationship between standard FoCUS views. The proposed framework uses a multi-view U-Net-like fully convolutional neural network to regress line-based heatmaps representing the most likely areas of intersection between input images. The lines that best fit the regressed heatmaps are then extracted, and a system of nonlinear equations based on the intersection between view triplets is created and solved to determine the relative 3D pose between all input images. The feasibility and accuracy of the proposed pipeline were validated using a novel realistic in silico FoCUS dataset, demonstrating promising results. Interestingly, as shown in preliminary experiments, the estimation of the 2D images' relative poses enables the application of 3D image analysis methods and paves the way for 3D quantitative assessments in FoCUS examinations.

Dual consistency loss for contour-aware segmentation in medical images.

Medical image segmentation is a paramount task for several clinical applications, namely for the diagnosis of pathologies, for treatment planning, and for aiding image-guided surgeries. With the development of deep learning, Convolutional Neural Networks (CNN) have become the state-of-the-art for medical image segmentation. However, issues are still raised concerning the precise object boundary delineation, since traditional CNNs can produce non-smooth segmentations with boundary discontinuities. In this work, a U-shaped CNN architecture is proposed to generate both pixel-wise segmentation and probabilistic contour maps of the object to segment, in order to generate reliable segmentations at the object's boundaries. Moreover, since the segmentation and contour maps must be inherently related to each other, a dual consistency loss that relates the two outputs of the network is proposed. Thus, the network is enforced to consistently learn the segmentation and contour delineation tasks during the training. The proposed method was applied and validated on a public dataset of cardiac 3D ultrasound images of the left ventricle. The results obtained showed the good performance of the method and its applicability for the cardiac dataset, showing its potential to be used in clinical practice for medical image segmentation.Clinical Relevance- The proposed network with dual consistency loss scheme can improve the performance of state-of-the-art CNNs for medical image segmentation, proving its value to be applied for computer-aided diagnosis.

Robust 3D breast reconstruction based on monocular images and artificial intelligence for robotic guided oncological interventions.

Breast cancer is a global public health concern. For women with suspicious breast lesions, the current diagnosis requires a biopsy, which is usually guided by ultrasound (US). However, this process is challenging due to the low quality of the US image and the complexity of dealing with the US probe and the surgical needle simultaneously, making it largely reliant on the surgeon's expertise. Some previous works employing collaborative robots emerged to improve the precision of biopsy interventions, providing an easier, safer, and more ergonomic procedure. However, for these equipment to be able to navigate around the breast autonomously, 3D breast reconstruction needs to be available. The accuracy of these systems still needs to improve, with the 3D reconstruction of the breast being one of the biggest focuses of errors. The main objective of this work is to develop a method to obtain a robust 3D reconstruction of the patient's breast, based on RGB monocular images, which later can be used to compute the robot's trajectories for the biopsy. To this end, depth estimation techniques will be developed, based on a deep learning architecture constituted by a CNN, LSTM, and MLP, to generate depth maps capable of being converted into point clouds. After merging several from multiple points of view, it is possible to generate a real-time reconstruction of the breast as a mesh. The development and validation of our method was performed using a previously described synthetic dataset. Hence, this procedure takes RGB images and the cameras' position and outputs the breasts' meshes. It has a mean error of 3.9 mm and a standard deviation of 1.2 mm. The final results attest to the ability of this methodology to predict the breast's shape and size using monocular images.Clinical Relevance- This work proposes a method based on artificial intelligence and monocular RGB images to obtain the breast's volume during robotic guided breast biopsies, improving their execution and safety.

Social determination of malaria in pregnancy in Colombia: a critical ethnographic study.

BACKGROUND: The meanings and experiences related to malaria in pregnancy (MiP) and its processes of social determination of health (PSDH) have not been reported in the world scientific literature. The objective was to understand the meanings and experiences of MiP, and to explain their PSDH in an endemic area from Colombia, 2022. METHODS: Critical ethnography with 46 pregnant women and 31 healthcare workers. In-depth and semi-structured interviews, focus group discussions, participant and non-participant observations, and field diaries were applied. A phenomenological-hermeneutic analysis, saturation and triangulation was carried out. The methodological rigor criteria were reflexivity, credibility, auditability, and transferability. RESULTS: At the singular level, participants indicated different problems in antenatal care and malaria control programmes, pregnant women were lacking knowledge about MiP, and malaria care was restricted to cases with high obstetric risk. Three additional levels that explain the PSDH of MiP were identified: (i) limitations of malaria control policies, and health-system, geographic, cultural and economic barriers by MiP diagnosis and treatment; (ii) problems of public health programmes and antenatal care; (iii) structural problems such as monetary poverty, scarcity of resources for public health and inefficiency in their use, lacking community commitment to preventive actions, and breach of institutional responsibilities of health promoter entity, municipalities and health services provider institutions. CONCLUSION: Initiatives for MiP control are concentrated at the singular level, PDSH identified in this research show the need to broaden the field of action, increase health resources, and improve public health programmes and antenatal care. It is also necessary to impact the reciprocal relationships of MiP with economic and cultural dimensions, although these aspects are increasingly diminished with the predominance and naturalization of neoliberal logic in health.

Fusion Object Detection and Action Recognition to Predict Violent Action.

In the context of Shared Autonomous Vehicles, the need to monitor the environment inside the car will be crucial. This article focuses on the application of deep learning algorithms to present a fusion monitoring solution which was three different algorithms: a violent action detection system, which recognizes violent behaviors between passengers, a violent object detection system, and a lost items detection system. Public datasets were used for object detection algorithms (COCO and TAO) to train state-of-the-art algorithms such as YOLOv5. For violent action detection, the MoLa InCar dataset was used to train on state-of-the-art algorithms such as I3D, R(2+1)D, SlowFast, TSN, and TSM. Finally, an embedded automotive solution was used to demonstrate that both methods are running in real-time.

CholecTriplet2022: Show me a tool and tell me the triplet - An endoscopic vision challenge for surgical action triplet detection.

Formalizing surgical activities as triplets of the used instruments, actions performed, and target anatomies is becoming a gold standard approach for surgical activity modeling. The benefit is that this formalization helps to obtain a more detailed understanding of tool-tissue interaction which can be used to develop better Artificial Intelligence assistance for image-guided surgery. Earlier efforts and the CholecTriplet challenge introduced in 2021 have put together techniques aimed at recognizing these triplets from surgical footage. Estimating also the spatial locations of the triplets would offer a more precise intraoperative context-aware decision support for computer-assisted intervention. This paper presents the CholecTriplet2022 challenge, which extends surgical action triplet modeling from recognition to detection. It includes weakly-supervised bounding box localization of every visible surgical instrument (or tool), as the key actors, and the modeling of each tool-activity in the form of triplet. The paper describes a baseline method and 10 new deep learning algorithms presented at the challenge to solve the task. It also provides thorough methodological comparisons of the methods, an in-depth analysis of the obtained results across multiple metrics, visual and procedural challenges; their significance, and useful insights for future research directions and applications in surgery.

A systematic review of mixed studies on malaria in Colombia 1980-2022: what the "bifocal vision" discovers.

Mixed methods are essential in public health research and malaria control, because they allow grasping part of the complexity and diversity of the factors that determine health-disease. This study analyzes the mixed studies on malaria in Colombia, 1980-2022, through a systematic review in 15 databases and institutional repositories. The methodological quality was assessed with Mixed Methods Appraisal Tool (MMAT), STrengthening the Reporting of OBservational studies in Epidemiology (STROBE), and Standards for Reporting Qualitative Research (SRQR). The qualitative and quantitative findings were grouped into a four-level hierarchical matrix. The epidemiological profile of malaria morbidity, from traditional epidemiology, has been sustained by environmental problems, armed conflict, individual risk behaviors, and low adherence to recommendations from health institutions. However, the qualitative component reveals deeper causes that are less studied, of greater theoretical complexity, and that reflect challenges to design and implement health interventions, such as socioeconomic and political crises, poverty, and the neoliberal orientation in the malaria control policy; the latter reflected in the change in the role of the State, the fragmentation of control actions, the predominance of insurance over social assistance, the privatization of the provision of health services, the individualistic and economistic predominance of health, and low connection with popular tradition and community initiatives. The above confirms the importance of expanding mixed studies as a source of evidence to improve malaria research and control models in Colombia, and to identify the underlying causes of the epidemiological profile.

Clinical and Parasitological Profiles of Gestational, Placental and Congenital Malaria in Northwestern Colombia.

This study compared the clinical-parasitological profiles of gestational (GM), placental (PM), and congenital (CM) malaria in northwestern Colombia. A cross-sectional study with 829 pregnant women, 549 placentas, and 547 newborns was conducted. The frequency of GM was 35.8%, PM 20.9%, and CM 8.5%. P. vivax predominated in GM; in PM, the proportion of P. vivax and P. falciparum was similar; in CM, P. falciparum predominated. The main clinical findings were headache (49%), anemia (32%), fever (24%), and musculoskeletal pain (13%). The clinical manifestations were statistically higher in P. vivax infections. In submicroscopic GM (positive with qPCR and negative with thick blood smear), the frequency of anemia, sore throat, and a headache was statistically higher compared with pregnant women without malaria. GM, PM, and CM reduce birth weight and head circumference. In Colombia, this is the first research on the clinical characteristics of GM, PM, and CM; contrary to evidence from other countries, P. vivax and submicroscopic infections are associated with clinical outcomes.

CholecTriplet2021: A benchmark challenge for surgical action triplet recognition.

Context-aware decision support in the operating room can foster surgical safety and efficiency by leveraging real-time feedback from surgical workflow analysis. Most existing works recognize surgical activities at a coarse-grained level, such as phases, steps or events, leaving out fine-grained interaction details about the surgical activity; yet those are needed for more helpful AI assistance in the operating room. Recognizing surgical actions as triplets of combination delivers more comprehensive details about the activities taking place in surgical videos. This paper presents CholecTriplet2021: an endoscopic vision challenge organized at MICCAI 2021 for the recognition of surgical action triplets in laparoscopic videos. The challenge granted private access to the large-scale CholecT50 dataset, which is annotated with action triplet information. In this paper, we present the challenge setup and the assessment of the state-of-the-art deep learning methods proposed by the participants during the challenge. A total of 4 baseline methods from the challenge organizers and 19 new deep learning algorithms from the competing teams are presented to recognize surgical action triplets directly from surgical videos, achieving mean average precision (mAP) ranging from 4.2% to 38.1%. This study also analyzes the significance of the results obtained by the presented approaches, performs a thorough methodological comparison between them, in-depth result analysis, and proposes a novel ensemble method for enhanced recognition. Our analysis shows that surgical workflow analysis is not yet solved, and also highlights interesting directions for future research on fine-grained surgical activity recognition which is of utmost importance for the development of AI in surgery.

Sensor Fusion Approach for Multiple Human Motion Detection for Indoor Surveillance Use-Case.

Multi-human detection and tracking in indoor surveillance is a challenging task due to various factors such as occlusions, illumination changes, and complex human-human and human-object interactions. In this study, we address these challenges by exploring the benefits of a low-level sensor fusion approach that combines grayscale and neuromorphic vision sensor (NVS) data. We first generate a custom dataset using an NVS camera in an indoor environment. We then conduct a comprehensive study by experimenting with different image features and deep learning networks, followed by a multi-input fusion strategy to optimize our experiments with respect to overfitting. Our primary goal is to determine the best input feature types for multi-human motion detection using statistical analysis. We find that there is a significant difference between the input features of optimized backbones, with the best strategy depending on the amount of available data. Specifically, under a low-data regime, event-based frames seem to be the preferred input feature type, while higher data availability benefits the combined use of grayscale and optical flow features. Our results demonstrate the potential of sensor fusion and deep learning techniques for multi-human tracking in indoor surveillance, although it is acknowledged that further studies are needed to confirm our findings.

Diagnostic Accuracy of a Thick Blood Smear Compared to qPCR for Malaria Associated with Pregnancy in Colombia.

This study aimed to evaluate the accuracy of the thick blood smear (TBS) versus quantitative polymerase chain reaction (qPCR) for the diagnosis of malaria associated with pregnancy (MAP) caused by P. falciparum or P. vivax in Colombia in its gestational malaria (GM), placental malaria (PM), and congenital malaria (CM) forms as well as to compare its accuracy in different subgroups of pregnant women according to the presence of fever, anemia and a history of malaria. This was a diagnostic evaluation of 829 pregnant women, 579 placentas, 381 umbilical cord samples, and 221 neonatal peripheral blood samples. Accuracy was evaluated based on the parameters of sensitivity, specificity, predictive values, likelihood ratios, and validity index, with their 95% confidence intervals. The frequency of GM was 36% (n = 297/829), PM 27% (n = 159/579), and CM 16.5% (n = 63/381) in umbilical cord samples and 2% (n = 5/221) in neonatal peripheral blood samples. For GM, the sensitivity was 55%, with higher rates in those infected with P. vivax (68%), with a history of malaria (69%), and with fever (96%). These three subgroups presented the best results in terms of the negative likelihood ratio and validity index. For PM, sensitivity was 8%; in subgroup analyses in terms of species, symptomatology (anemia and fever), and history of malaria, it was 1-18%, and the negative likelihood ratio was >0.80 in all subgroups. No false positives were recorded in any of the subgroups. The TBS did not detect any cases of CM. This study found the TBS yielded satisfactory results in terms of diagnosing GM for P. vivax, pregnant women with previous malaria and febrile. It also showed that the TBS is not useful for diagnosing PM and CM. It is necessary to conduct surveillance of MAP with molecular methods in in groups where TBS is deficient (asymptomatic GM, P. falciparum, and pregnant women without history of malaria) to optimize the timely treatment of PM and CM, avoid the deleterious effects of MAP and achieve the malaria elimination goals in Colombia.

A multi-task convolutional neural network for classification and segmentation of chronic venous disorders.

Chronic Venous Disorders (CVD) of the lower limbs are one of the most prevalent medical conditions, affecting 35% of adults in Europe and North America. Due to the exponential growth of the aging population and the worsening of CVD with age, it is expected that the healthcare costs and the resources needed for the treatment of CVD will increase in the coming years. The early diagnosis of CVD is fundamental in treatment planning, while the monitoring of its treatment is fundamental to assess a patient's condition and quantify the evolution of CVD. However, correct diagnosis relies on a qualitative approach through visual recognition of the various venous disorders, being time-consuming and highly dependent on the physician's expertise. In this paper, we propose a novel automatic strategy for the joint segmentation and classification of CVDs. The strategy relies on a multi-task deep learning network, denominated VENet, that simultaneously solves segmentation and classification tasks, exploiting the information of both tasks to increase learning efficiency, ultimately improving their performance. The proposed method was compared against state-of-the-art strategies in a dataset of 1376 CVD images. Experiments showed that the VENet achieved a classification performance of 96.4%, 96.4%, and 97.2% for accuracy, precision, and recall, respectively, and a segmentation performance of 75.4%, 76.7.0%, 76.7% for the Dice coefficient, precision, and recall, respectively. The joint formulation increased the robustness of both tasks when compared to the conventional classification or segmentation strategies, proving its added value, mainly for the segmentation of small lesions.

Systematic review of the diagnostic accuracy of thick smear compared to polymerase chain reaction for pregnancy-associated malaria, 2010-2022. / Revisión sistemática de la exactitud diagnóstica de la gota gruesa en comparación con la reacción en cadena de la polimerasa para malaria asociada al embarazo, 2010-2022.

OBJECTIVE.: To evaluate the accuracy of thick smear (TS) versus quantitative polymerase chain reaction (PCR) for pregnancy-associated malaria (PAM). MATERIALS AND METHODS.: We carried out a systematic review of diagnostic tests in nine databases. Methodological quality was evaluated with QUADAS. Sensitivity, specificity, positive likelihood ratio (PLR), negative likelihood ratio (NLR), diagnostic odds ratio (DOR) and area under the ROC curve were estimated. Heterogeneity was determined with the Der Simonian-Laird Q method and uncertainty with the weighted percentage of each study on the overall result. RESULTS.: We included 10 studies with 5691 pregnant women, 1415 placentas and 84 neonates. In the studies with nested PCR (nPCR) and quantitative PCR (qPCR) as the standard, the diagnostic accuracy results were statistically similar, with very low sensitivity (50 and 54%, respectively), high specificity (99% in both cases), high PLR and poor NLR. When nPCR was used, the DOR was 162 (95%CI=66-401) and the area under the ROC curve was 95%, while with qPCR it was 231 (95%CI=27-1951) and 78%, respectively. CONCLUSIONS.: We demonstrated that research on the diagnostic accuracy of TS in PAM is limited. Microscopy showed poor performance in the diagnosis of asymptomatic or low parasitemia infections, which reinforces the importance of implementing other types of techniques for the follow-up and control of malaria infections in pregnant women, in order to achieve the control and possible elimination of PAM.

OBJETIVOS.: Evaluar la exactitud de gota gruesa (GG) frente a la reacción en cadena de la polimerasa (PCR) cuantitativa para la malaria asociada al embarazo (MAE). MATERIALES Y MÉTODOS.: Se realizó una revisión sistemática de pruebas diagnósticas en nueve bases de datos. Se evaluó la calidad metodológica con QUADAS. Se estimó sensibilidad, especificidad, cociente de probabilidad positivo (CPP) y negativo (CPN), razón de odds diagnóstica (ORD) y área bajo la curva ROC. Se determinó la heterogeneidad con el estadístico Q de Der Simonian-Laird y la incertidumbre con el porcentaje de peso de cada estudio sobre el resultado global. RESULTADOS.: Se incluyeron diez estudios con 5691 gestantes, 1415 placentas y 84 neonatos. En los estudios con nPCR (PCR anidada) y qPCR (PCR cuantitativa) como estándar, los resultados de exactitud diagnóstica fueron estadísticamente similares, con sensibilidad muy baja (50 y 54%, respectivamente), alta especificidad (99% en ambos casos), alto CPP y deficiente CPN. Usando nPCR la OR diagnóstica fue 162 (IC95%=66-401) y el área bajo la curva ROC fue 95%, mientras que con qPCR fueron 231 (IC95%=27-1951) y 78%, respectivamente. CONCLUSIONES.: Mediante un protocolo exhaustivo se demostró el bajo desarrollo de investigaciones sobre la exactitud diagnóstica de la GG en MAE. Se demostró que la microscopía tiene un desempeño deficiente para el diagnóstico de infecciones asintomáticas o de baja parasitemia, lo que afianza la importancia de implementar otro tipo de técnicas en el seguimiento y control de las infecciones por malaria en las gestantes, con el fin de lograr el control y posible eliminación de la MAE.

First colombian tricuspid endovascular valve in valve using kissing balloon technique for failed bioprosthesis annular cracking / El primer caso colombiano de reemplazo percutáneo válvula en válvula tricúspide utilizando la técnica de kissing balloon con fractura anular por falla bioprostética

Abstract We present the first case in Colombia of tricuspid endovascular valve in valve for failed bioprosthesis in a 40 years old patient with very high operative risk with great results, proposing kissing balloon annulus cracking technique as a practical solution for the colombian specialists.

Resumen Se presenta el primer caso en Colombia de un reemplazo percutáneo tipo válvula en válvula por falla de bioprótesis tricúspide en un paciente de 40 años con un muy alto riesgo quirúrgico, con excelentes resultados, proponiendo la técnica kissing balloon de fractura anular como una solución práctica para los especialistas colombianos.

Characterization of the Workspace and Limits of Operation of Laser Treatments for Vascular Lesions of the Lower Limbs.

The increase of the aging population brings numerous challenges to health and aesthetic segments. Here, the use of laser therapy for dermatology is expected to increase since it allows for non-invasive and infection-free treatments. However, existing laser devices require doctors' manually handling and visually inspecting the skin. As such, the treatment outcome is dependent on the user's expertise, which frequently results in ineffective treatments and side effects. This study aims to determine the workspace and limits of operation of laser treatments for vascular lesions of the lower limbs. The results of this study can be used to develop a robotic-guided technology to help address the aforementioned problems. Specifically, workspace and limits of operation were studied in eight vascular laser treatments. For it, an electromagnetic tracking system was used to collect the real-time positioning of the laser during the treatments. The computed average workspace length, height, and width were 0.84 ± 0.15, 0.41 ± 0.06, and 0.78 ± 0.16 m, respectively. This corresponds to an average volume of treatment of 0.277 ± 0.093 m3. The average treatment time was 23.2 ± 10.2 min, with an average laser orientation of 40.6 ± 5.6 degrees. Additionally, the average velocities of 0.124 ± 0.103 m/s and 31.5 + 25.4 deg/s were measured. This knowledge characterizes the vascular laser treatment workspace and limits of operation, which may ease the understanding for future robotic system development.

Effect of Plasmodium Infection during Pregnancy on Passive Neonatal Immunity against Tetanus Toxoid and Rotavirus.

Passive immunity acquired through transplacental IgG transport is essential to protect infants against pathogens as childhood vaccination programs begins. Diarrhea caused by rotavirus and neonatal tetanus are common and potentially fatal childhood infections that can be prevented by transplacental IgG. However, it is not known whether maternal infections in pregnancy can reduce the transfer of these antibodies to the fetus. This study evaluated the effect of submicroscopic Plasmodium infection during pregnancy on the transfer of maternal IgG antibodies against rotavirus (anti-RV) and tetanus toxoid (anti-TT) to newborns of pregnant women residing in Puerto Libertador and Tierralta, Colombia. Expression of different immune mediators and levels of IgG against rotavirus and tetanus toxoid were quantified in pregnant women with and without Plasmodium infection during pregnancy. Submicroscopic infection at the time of delivery was associated with a cord-to-maternal ratio (CMR) > 1 for anti-RV and < 1 for anti-TT IgG, as well as with an increase in the expression of immune mediators of inflammation (IFN-γ), anti-inflammation (IL-10, TGF-ß), and regulation (FoxP3, CTLA-4). When compared by species, these findings (CMR > 1 for anti-RV and < 1 for anti-TT IgG) were conserved in submicroscopic Plasmodium vivax infections at delivery. The impact of Plasmodium infections on neonatal susceptibility to other infections warrants further exploration.

AI based monitoring violent action detection data for in-vehicle scenarios.

With the evolution of technology associated with mobility and autonomy, Shared Autonomous Vehicles will be a reality. To ensure passenger safety, there is a need to create a monitoring system inside the vehicle capable of recognizing human actions. We introduce two datasets to train human action recognition inside the vehicle, focusing on violence detection. The InCar dataset tackles violent actions for in-car background which give us more realistic data. The InVicon dataset although doesn't have the realistic background as the InCar dataset can provide skeleton (3D body joints) data. This datasets were recorded with RGB, Depth, Thermal, Event-based, and Skeleton data. The resulting dataset contains 6 400 video samples and more than 3 million frames, collected from sixteen distinct subjects. The dataset contains 58 action classes, including violent and neutral (i.e., non-violent) activities.

3D Facial Landmark Localization for cephalometric analysis.

Cephalometric analysis is an important and routine task in the medical field to assess craniofacial development and to diagnose cranial deformities and midline facial abnormalities. The advance of 3D digital techniques potentiated the development of 3D cephalometry, which includes the localization of cephalometric landmarks in the 3D models. However, manual labeling is still applied, being a tedious and time-consuming task, highly prone to intra/inter-observer variability. In this paper, a framework to automatically locate cephalometric landmarks in 3D facial models is presented. The landmark detector is divided into two stages: (i) creation of 2D maps representative of the 3D model; and (ii) landmarks' detection through a regression convolutional neural network (CNN). In the first step, the 3D facial model is transformed to 2D maps retrieved from 3D shape descriptors. In the second stage, a CNN is used to estimate a probability map for each landmark using the 2D representations as input. The detection method was evaluated in three different datasets of 3D facial models, namely the Texas 3DFR, the BU3DFE, and the Bosphorus databases. An average distance error of 2.3, 3.0, and 3.2 mm were obtained for the landmarks evaluated on each dataset. The obtained results demonstrated the accuracy of the method in different 3D facial datasets with a performance competitive to the state-of-the-art methods, allowing to prove its versability to different 3D models. Clinical Relevance- Overall, the performance of the landmark detector demonstrated its potential to be used for 3D cephalometric analysis.

Classification of Chronic Venous Disorders using an Ensemble Optimization of Convolutional Neural Networks.

Chronic Venous Disorders (CVD) of lower limbs are one of the most prevalent medical conditions, affecting 35% of adults in Europe and North America. The early diagnosis of CVD is critical, however, the diagnosis relies on a visual recognition of the various venous disorders which is time- consuming and dependent on the physician's expertise. Thus, automatic strategies for the classification of the CVD severity are claimed. This paper proposed an automatic ensemble-based strategy of Deep Convolutional Neural Networks (DCNN) for the classification of CVDs severity from medical images. First, a clinical dataset containing 1376 images of patients' legs with CVD of 5 different levels of severity was constructed. Then, the constructed dataset was randomly split into training, testing, and validation datasets. Subsequently, a set of DCNN were individually applied to the images for classification. Finally, instead of a traditional voting ensemble strategy, extracted feature vectors from each DCNN were concatenated and fed into a new ensemble optimization network. Experiments showed that the proposed strategy achieved a classification with 93.8%, 93.4%, 92.4% of accuracy, precision, and recall, respectively. Moreover, compared to the traditional ensemble strategy, improvement in the accuracy of ~2% was registered. The proposed strategy showed to be accurate and robust for the diagnosis of CVD severity from medical images. Nevertheless, further research using an extensive clinical database is required to validate the potential of this strategy. Clinical Relevance- An automatic classification of CVD to reduce the probability of underdiagnoses and promote the treatment of CVD in the early stages.