DNA damage targeted therapy for advanced breast cancer.

Breast cancer (BC) is the most prevalent malignancy affecting women worldwide, including Portugal. While the majority of BC cases are sporadic, hereditary forms account for 5-10% of cases. The most common inherited mutations associated with BC are germline mutations in the BReast CAncer (BRCA) 1/2 gene (gBRCA1/2). They are found in approximately 5-6% of BC patients and are inherited in an autosomal dominant manner, primarily affecting younger women. Pathogenic variants within BRCA1/2 genes elevate the risk of both breast and ovarian cancers and give rise to distinct clinical phenotypes. BRCA proteins play a key role in maintaining genome integrity by facilitating the repair of double-strand breaks through the homologous recombination (HR) pathway. Therefore, any mutation that impairs the function of BRCA proteins can result in the accumulation of DNA damage, genomic instability, and potentially contribute to cancer development and progression. Testing for gBRCA1/2 status is relevant for treatment planning, as it can provide insights into the likely response to therapy involving platinum-based chemotherapy and poly[adenosine diphosphate (ADP)-ribose] polymerase inhibitors (PARPi). The aim of this review was to investigate the impact of HR deficiency in BC, focusing on BRCA mutations and their impact on the modulation of responses to platinum and PARPi therapy, and to share the experience of Unidade Local de Saúde Santa Maria in the management of metastatic BC patients with DNA damage targeted therapy, including those with the Portuguese c.156_157insAlu BRCA2 founder mutation.

Deep learning models for interpretation of point of care ultrasound in military working dogs.

Introduction: Military working dogs (MWDs) are essential for military operations in a wide range of missions. With this pivotal role, MWDs can become casualties requiring specialized veterinary care that may not always be available far forward on the battlefield. Some injuries such as pneumothorax, hemothorax, or abdominal hemorrhage can be diagnosed using point of care ultrasound (POCUS) such as the Global FAST® exam. This presents a unique opportunity for artificial intelligence (AI) to aid in the interpretation of ultrasound images. In this article, deep learning classification neural networks were developed for POCUS assessment in MWDs. Methods: Images were collected in five MWDs under general anesthesia or deep sedation for all scan points in the Global FAST® exam. For representative injuries, a cadaver model was used from which positive and negative injury images were captured. A total of 327 ultrasound clips were captured and split across scan points for training three different AI network architectures: MobileNetV2, DarkNet-19, and ShrapML. Gradient class activation mapping (GradCAM) overlays were generated for representative images to better explain AI predictions. Results: Performance of AI models reached over 82% accuracy for all scan points. The model with the highest performance was trained with the MobileNetV2 network for the cystocolic scan point achieving 99.8% accuracy. Across all trained networks the diaphragmatic hepatorenal scan point had the best overall performance. However, GradCAM overlays showed that the models with highest accuracy, like MobileNetV2, were not always identifying relevant features. Conversely, the GradCAM heatmaps for ShrapML show general agreement with regions most indicative of fluid accumulation. Discussion: Overall, the AI models developed can automate POCUS predictions in MWDs. Preliminarily, ShrapML had the strongest performance and prediction rate paired with accurately tracking fluid accumulation sites, making it the most suitable option for eventual real-time deployment with ultrasound systems. Further integration of this technology with imaging technologies will expand use of POCUS-based triage of MWDs.

A scoping review to create a framework for the steps in developing condition-specific preference-based instruments de novo or from an existing non-preference-based instrument: use of item response theory or Rasch analysis.

BACKGROUND: There is no widely accepted framework to guide the development of condition-specific preference-based instruments (CSPBIs) that includes both de novo and from existing non-preference-based instruments. The purpose of this study was to address this gap by reviewing the published literature on CSPBIs, with particular attention to the application of item response theory (IRT) and Rasch analysis in their development. METHODS: A scoping review of the literature covering the concepts of all phases of CSPBI development and evaluation was performed from MEDLINE, Embase, PsychInfo, CINAHL, and the Cochrane Library, from inception to December 30, 2022. RESULTS: The titles and abstracts of 1,967 unique references were reviewed. After retrieving and reviewing 154 full-text articles, data were extracted from 109 articles, representing 41 CSPBIs covering 21 diseases or conditions. The development of CSPBIs was conceptualized as a 15-step framework, covering four phases: 1) develop initial questionnaire items (when no suitable non-preference-based instrument exists), 2) establish the dimensional structure, 3) reduce items per dimension, 4) value and model health state utilities. Thirty-nine instruments used a type of Rasch model and two instruments used IRT models in phase 3. CONCLUSION: We present an expanded framework that outlines the development of CSPBIs, both from existing non-preference-based instruments and de novo when no suitable non-preference-based instrument exists, using IRT and Rasch analysis. For items that fit the Rasch model, developers selected one item per dimension and explored item response level reduction. This framework will guide researchers who are developing or assessing CSPBIs.

Active Bone Conduction Implant and Adhesive Bone Conduction Device: A Comparison of Audiological Performance and Subjective Satisfaction.

Introduction Atresia of the external auditory canal affects 1 in every 10 thousand to 20 thousand live births, with a much higher prevalence in Latin America, at 5 to 21 out of every 10 thousand newborns. The treatment involves esthetic and functional aspects. Regarding the functional treatment, there are surgical and nonsurgical alternatives like spectacle frames and rigid and softband systems. Active transcutaneous bone conduction implants (BCIs) achieve good sound transmission and directly stimulate the bone. Objective To assess the audiological performance and subjective satisfaction of children implanted with an active transcutaneous BCI for more than one year and to compare the outcomes with a nonsurgical adhesive bone conduction device (aBCD) in the same users. Methods The present is a prospective, multicentric study. The audiological performance was evaluated at 1, 6, and 12 months postactivation, and after a 1-month trial with the nonsurgical device. Results Ten patients completed all tests. The 4-frequency pure-tone average (4PTA) in the unaided condition was of 65 dB HL, which improved significantly to 20 dB HL after using the BCI for 12 months. The speech recognition in quiet in the unaided condition was of 33% on average, which improved significantly, to 99% with the BCI, and to 91% with the aBCD. Conclusion The aBCD demonstrated sufficient hearing improvement and subjective satisfaction; thus, it is a good solution for hearing rehabilitation if surgery is not desired or not possible. If surgery is an option, the BCI is the superior device in terms of hearing outcomes, particularly background noise and subjective satisfaction.

Evaluation of Deep Learning Model Architectures for Point-of-Care Ultrasound Diagnostics.

Point-of-care ultrasound imaging is a critical tool for patient triage during trauma for diagnosing injuries and prioritizing limited medical evacuation resources. Specifically, an eFAST exam evaluates if there are free fluids in the chest or abdomen but this is only possible if ultrasound scans can be accurately interpreted, a challenge in the pre-hospital setting. In this effort, we evaluated the use of artificial intelligent eFAST image interpretation models. Widely used deep learning model architectures were evaluated as well as Bayesian models optimized for six different diagnostic models: pneumothorax (i) B- or (ii) M-mode, hemothorax (iii) B- or (iv) M-mode, (v) pelvic or bladder abdominal hemorrhage and (vi) right upper quadrant abdominal hemorrhage. Models were trained using images captured in 27 swine. Using a leave-one-subject-out training approach, the MobileNetV2 and DarkNet53 models surpassed 85% accuracy for each M-mode scan site. The different B-mode models performed worse with accuracies between 68% and 74% except for the pelvic hemorrhage model, which only reached 62% accuracy for all model architectures. These results highlight which eFAST scan sites can be easily automated with image interpretation models, while other scan sites, such as the bladder hemorrhage model, will require more robust model development or data augmentation to improve performance. With these additional improvements, the skill threshold for ultrasound-based triage can be reduced, thus expanding its utility in the pre-hospital setting.

Design and testing of ultrasound probe adapters for a robotic imaging platform.

Medical imaging-based triage is a critical tool for emergency medicine in both civilian and military settings. Ultrasound imaging can be used to rapidly identify free fluid in abdominal and thoracic cavities which could necessitate immediate surgical intervention. However, proper ultrasound image capture requires a skilled ultrasonography technician who is likely unavailable at the point of injury where resources are limited. Instead, robotics and computer vision technology can simplify image acquisition. As a first step towards this larger goal, here, we focus on the development of prototypes for ultrasound probe securement using a robotics platform. The ability of four probe adapter technologies to precisely capture images at anatomical locations, repeatedly, and with different ultrasound transducer types were evaluated across more than five scoring criteria. Testing demonstrated two of the adapters outperformed the traditional robot gripper and manual image capture, with a compact, rotating design compatible with wireless imaging technology being most suitable for use at the point of injury. Next steps will integrate the robotic platform with computer vision and deep learning image interpretation models to automate image capture and diagnosis. This will lower the skill threshold needed for medical imaging-based triage, enabling this procedure to be available at or near the point of injury.

Toward Smart, Automated Junctional Tourniquets-AI Models to Interpret Vessel Occlusion at Physiological Pressure Points.

Hemorrhage is the leading cause of preventable death in both civilian and military medicine. Junctional hemorrhages are especially difficult to manage since traditional tourniquet placement is often not possible. Ultrasound can be used to visualize and guide the caretaker to apply pressure at physiological pressure points to stop hemorrhage. However, this process is technically challenging, requiring the vessel to be properly positioned over rigid boney surfaces and applying sufficient pressure to maintain proper occlusion. As a first step toward automating this life-saving intervention, we demonstrate an artificial intelligence algorithm that classifies a vessel as patent or occluded, which can guide a user to apply the appropriate pressure required to stop flow. Neural network models were trained using images captured from a custom tissue-mimicking phantom and an ex vivo swine model of the inguinal region, as pressure was applied using an ultrasound probe with and without color Doppler overlays. Using these images, we developed an image classification algorithm suitable for the determination of patency or occlusion in an ultrasound image containing color Doppler overlay. Separate AI models for both test platforms were able to accurately detect occlusion status in test-image sets to more than 93% accuracy. In conclusion, this methodology can be utilized for guiding and monitoring proper vessel occlusion, which, when combined with automated actuation and other AI models, can allow for automated junctional tourniquet application.

Using AI Segmentation Models to Improve Foreign Body Detection and Triage from Ultrasound Images.

Medical imaging can be a critical tool for triaging casualties in trauma situations. In remote or military medicine scenarios, triage is essential for identifying how to use limited resources or prioritize evacuation for the most serious cases. Ultrasound imaging, while portable and often available near the point of injury, can only be used for triage if images are properly acquired, interpreted, and objectively triage scored. Here, we detail how AI segmentation models can be used for improving image interpretation and objective triage evaluation for a medical application focused on foreign bodies embedded in tissues at variable distances from critical neurovascular features. Ultrasound images previously collected in a tissue phantom with or without neurovascular features were labeled with ground truth masks. These image sets were used to train two different segmentation AI frameworks: YOLOv7 and U-Net segmentation models. Overall, both approaches were successful in identifying shrapnel in the image set, with U-Net outperforming YOLOv7 for single-class segmentation. Both segmentation models were also evaluated with a more complex image set containing shrapnel, artery, vein, and nerve features. YOLOv7 obtained higher precision scores across multiple classes whereas U-Net achieved higher recall scores. Using each AI model, a triage distance metric was adapted to measure the proximity of shrapnel to the nearest neurovascular feature, with U-Net more closely mirroring the triage distances measured from ground truth labels. Overall, the segmentation AI models were successful in detecting shrapnel in ultrasound images and could allow for improved injury triage in emergency medicine scenarios.

Pulmonary valve replacement in tetralogy of Fallot - who and how?

BACKGROUND AND AIM: Pulmonary regurgitation is the most common complication in repaired tetralogy of Fallot patients. Severe chronic pulmonary regurgitation can be tolerated for decades, but if not treated, it can progress to symptomatic, irreversible right ventricular dilatation and dysfunction. We investigated clinical associations with pulmonary valve replacement among patients with significative pulmonary regurgitation and how interventional developments can change their management. METHODS: All adult patients with repaired tetralogy of Fallot who were followed at an adult CHD Clinic at a single centre from 1980 to 2022 were included on their first outpatient visit. Follow-up was estimated from the time of correction surgery until one of the following events occurred first: pulmonary valve replacement, death, loss to follow-up or conclusion of the study. RESULTS: We included 221 patients (116 males) with a median age of 19 (18-25). At a median age of 33 (10) years old, 114 (51%) patients presented significant pulmonary regurgitation. Among patients with significant pulmonary regurgitation, pulmonary valve replacement was associated with male gender, older age at surgical repair, and longer QRS duration in adulthood. Pulmonary valve replacement was performed in 50 patients, including four transcatheter pulmonary valve implantations, at a median age of 34 (14) years. CONCLUSION: Pulmonary regurgitation affects a large percentage of tetralogy of Fallot adult patients, requiring a long-term clinical and imaging follow-up. Sex, age at surgical repair and longer QRS are associated with the need of PVR among patients with significative pulmonary regurgitation. Clinical practice and current literature support TPVI as the future gold standard intervention.

Active Bone Conduction Implant and Adhesive Bone Conduction Device: A Comparison of Audiological Performance and Subjective Satisfaction

Abstract Introduction Atresia of the external auditory canal affects 1 in every 10 thousand to 20 thousand live births, with a much higher prevalence in Latin America, at 5 to 21 out of every 10 thousand newborns. The treatment involves esthetic and functional aspects. Regarding the functional treatment, there are surgical and nonsurgical alternatives like spectacle frames and rigid and softband systems. Active transcutaneous bone conduction implants (BCIs) achieve good sound transmission and directly stimulate the bone. Objective To assess the audiological performance and subjective satisfaction of children implanted with an active transcutaneous BCI for more than one year and to compare the outcomes with a nonsurgical adhesive bone conduction device (aBCD) in the same users. Methods The present is a prospective, multicentric study. The audiological performance was evaluated at 1, 6, and 12 months postactivation, and after a 1-month trial with the nonsurgical device. Results Ten patients completed all tests. The 4-frequency pure-tone average (4PTA) in the unaided condition was of 65 dB HL, which improved significantly to 20 dB HL after using the BCI for 12 months. The speech recognition in quiet in the unaided condition was of 33% on average, which improved significantly, to 99% with the BCI, and to 91% with the aBCD. Conclusion The aBCD demonstrated sufficient hearing improvement and subjective satisfaction; thus, it is a good solution for hearing rehabilitation if surgery is not desired or not possible. If surgery is an option, the BCI is the superior device in terms of hearing outcomes, particularly background noise and subjective satisfaction.

[Numerical expression of the clinical course of the disease. Data management]. / Expresión numérica del curso clínico de la enfermedad. Manejo de datos.

Data management "behind the scenes" refers to collection, cleaning, imputation, and demarcation; and despite of being indispensable processes, they are usually neglected and thus, generate erroneous information. During the collection are errors: omission of covariates, deviation from the objective, and insufficient quality. The omission of covariates distorts the result attributed to the main manoeuvre. Deviation from the primary objective commonly occurs when the outcome is rare, delayed, or subjective and promotes substitution by non-equivalent surrogate variables. Moreover, insufficient quality occurs due to inadequate instruments, omission of the measurement procedure, or measurements out of context, such as attribution at the wrong time or equivalent. Furthermore, cleaning implies identifying erroneous, extreme, and missing values, which may or may not be imputed, depending on the percentage. The values of the manoeuvre or the outcome are never imputed, nor are patients eliminated due to a lack of values. Finally, the demarcation of each variable seeks to give it a clinical meaning about the outcome, for which a hierarchical sequence of criteria is followed: 1) previous clinical study, 2) expert agreement, 3) clinical judgment of the investigator/investigators, and 4) statistics. Acting without quality controls in data management frequently causes involuntary lies and confuses instead of clarifying.

El manejo de datos "tras bambalinas" se refiere a los procesos de recopilación, limpieza, imputación y demarcación; los cuales, aun siendo indispensables, usualmente suelen ser descuidados, por lo que generan información errónea. Durante la recopilación son errores: omisión de covariables, desvío del objetivo, y calidad insuficiente. La omisión de covariables distorsiona el resultado atribuido a la maniobra principal. El desvío del objetivo primario es común cuando el desenlace es raro, tardado o subjetivo y promueve la sustitución por variables subrogadas no equivalentes. Además, la calidad insuficiente, sucede por instrumentos inadecuados, omisión del procedimiento de medición, o medición fuera de contexto -como atribución a destiempo o equivalente-. Por otro lado, la limpieza implica identificar valores erróneos, extremos y faltantes, que podrán ser o no imputados, dependiendo del porcentaje se imputará comúnmente por la medida de resumen. Nunca se imputan los valores de la maniobra ni del desenlace, ni se eliminan pacientes por falta de valores. Finalmente, la demarcación de cada variable busca un significado clínico en referencia al desenlace, para ello se sigue una secuencia jerárquica de criterios: 1) estudio clínico previo, 2) acuerdo de expertos, 3) juicio clínico del investigador/investigadores y 4) estadística. Actuar sin controles de calidad en el manejo de datos provoca frecuentemente mentiras involuntarias y confunde en lugar de esclarecer.

An extended focused assessment with sonography in trauma ultrasound tissue-mimicking phantom for developing automated diagnostic technologies.

Introduction: Medical imaging-based triage is critical for ensuring medical treatment is timely and prioritized. However, without proper image collection and interpretation, triage decisions can be hard to make. While automation approaches can enhance these triage applications, tissue phantoms must be developed to train and mature these novel technologies. Here, we have developed a tissue phantom modeling the ultrasound views imaged during the enhanced focused assessment with sonography in trauma exam (eFAST). Methods: The tissue phantom utilized synthetic clear ballistic gel with carveouts in the abdomen and rib cage corresponding to the various eFAST scan points. Various approaches were taken to simulate proper physiology without injuries present or to mimic pneumothorax, hemothorax, or abdominal hemorrhage at multiple locations in the torso. Multiple ultrasound imaging systems were used to acquire ultrasound scans with or without injury present and were used to train deep learning image classification predictive models. Results: Performance of the artificial intelligent (AI) models trained in this study achieved over 97% accuracy for each eFAST scan site. We used a previously trained AI model for pneumothorax which achieved 74% accuracy in blind predictions for images collected with the novel eFAST tissue phantom. Grad-CAM heat map overlays for the predictions identified that the AI models were tracking the area of interest for each scan point in the tissue phantom. Discussion: Overall, the eFAST tissue phantom ultrasound scans resembled human images and were successful in training AI models. Tissue phantoms are critical first steps in troubleshooting and developing medical imaging automation technologies for this application that can accelerate the widespread use of ultrasound imaging for emergency triage.

Verification and Validation of Lower Body Negative Pressure as a Non-Invasive Bioengineering Tool for Testing Technologies for Monitoring Human Hemorrhage.

Since hemorrhage is a leading cause of preventable death in both civilian and military settings, the development of advanced decision support monitoring capabilities is necessary to promote improved clinical outcomes. The emergence of lower body negative pressure (LBNP) has provided a bioengineering technology for inducing progressive reductions in central blood volume shown to be accurate as a model for the study of the early compensatory stages of hemorrhage. In this context, the specific aim of this study was to provide for the first time a systematic technical evaluation to meet a commonly accepted engineering standard based on the FDA-recognized Standard for Assessing Credibility of Modeling through Verification and Validation (V&V) for Medical Devices (ASME standard V&V 40) specifically highlighting LBNP as a valuable resource for the safe study of hemorrhage physiology in humans. As an experimental tool, evidence is presented that LBNP is credible, repeatable, and validated as an analog for the study of human hemorrhage physiology compared to actual blood loss. The LBNP tool can promote the testing and development of advanced monitoring algorithms and evaluating wearable sensors with the goal of improving clinical outcomes during use in emergency medical settings.

Immunotherapy around the Clock: Impact of Infusion Timing on Stage IV Melanoma Outcomes.

Although the impact of circadian timing on immunotherapy has yet to be integrated into clinical practice, chronoimmunotherapy is an emerging and promising field as circadian oscillations are observed in immune cell numbers as well as the expression of immunotherapy targets, e.g., programmed cell death protein-1 and its ligand programmed death ligand 1. Concurrent retrospective studies suggest that morning infusions may lead to higher effectiveness of immune checkpoint inhibitors in melanoma, non-small cell lung cancer, and kidney cancer. This paper discusses the results of a retrospective study (2016-2022) exploring the impact of infusion timing on the outcomes of all 73 patients with stage IV melanoma receiving immunotherapy at a particular medical center. While the median overall survival (OS) was 24.2 months (95% confidence interval [CI] 9.04-39.8), for a median follow-up of 15.3 months, our results show that having more than 75% of infusions in the afternoon results in shorter median OS (14.9 vs. 38.1 months; hazard ratio 0.45 [CI 0.23-0.86]; p < 0.01) with more expressive impacts on particular subgroups: women, older patients, and patients with a lower tumor burden at the outset of immunotherapy. Our findings highlight the potential benefits of follow-up validation in prospective and translational randomized studies.

Breast Sarcomas, Phyllodes Tumors, and Desmoid Tumors: Turning the Magnifying Glass on Rare and Aggressive Entities.

Breast sarcomas (BSs), phyllodes tumors (PTs), and desmoid tumors (DTs) are rare entities that arise from connective tissue. BSs can be classified as either primary or secondary, whether they develop de novo or after radiation exposure or lymphedema. PIK3CA seems to play an important common role in different BS. Malignant PTs show similar behavior to BSs, while DTs are locally aggressive but rarely metastasize. BSs usually present as unilateral, painless, rapidly growing masses with rare nodal involvement. The diagnosis should be based on magnetic resonance imaging and a core needle biopsy. Staging should comprise a chest computed tomography (CT) scan (except for benign PT and DT), while abdominal and pelvic CT scans and bone scans should be added in certain subtypes. The mainstay of treatment for localized BS is surgery, with margin goals that vary according to subtype. Radiotherapy and chemotherapy can be used as neoadjuvant or adjuvant approaches, but their use in these settings is not standard. Advanced BS should be treated with systemic therapy, consistent with recommendations for advanced soft tissue sarcomas of other topographies. Given the rarity and heterogeneity of these entities, multidisciplinary and multi-institutional collaboration and treatment at reference centers are critical.

[Risk management system to reduce COVID-19 infection in health personnelSistema de gestão de riscos para reduzir o contágio de COVID-19 em profissionais de saúde]. / Sistema de gestión de riesgos para reducir el contagio de COVID-19 en el personal de salud.

Objective: To assess the occupational factors associated with the occurrence of COVID-19 in health personnel who were exposed to different magnitudes of risk and who followed the United Nations crisis management policy for COVID-19. Methods: Cross-sectional survey conducted between April and May 2021. The low-risk group (LRG) were considered to be those who had minimal contact with patients; the medium-risk group (MRG) had contact with non-COVID-19 patients and did not perform instrumental airway intervention; and the high-risk group (HRG) were those who cared for COVID-19 patients and performed instrumental intervention with aerosol generation. Diagnosed COVID-19 disease and the presence of positive IgG antibodies for SARS-CoV-2 measured with Elecsys® anti-SARS-CoV-2 were considered as outcomes. Results: Outcome recorded in 43.8% of the LRG, versus 46.7% in the MRG (odds ratio [OR]: 1.125; 95% confidence interval [CI 95% ]: 0.896-1.414; p = 0.311), and 48.6% in the HRG (OR: 1,214; CI 95%: 0.964-1.530; p= 0.10). Conclusion: Belonging to the high-risk group and the medium-risk group, based on the degree of exposure to confirmed COVID-19 patients in the work area, was not associated with a higher occurrence of disease or seroconversion.

Objetivo: Avaliar os fatores ocupacionais associados à ocorrência de COVID-19 em profissionais de saúde expostos a diferentes níveis de risco utilizando a política de gestão de crises elaborada pelas Nações Unidas para a COVID-19. Métodos: Pesquisa transversal realizada entre abril e maio de 2021. O grupo de risco baixo (GRB) consistia em profissionais que tinham contato mínimo com os pacientes; o grupo de risco médio (GRM) incluía profissionais que tinham contato com pacientes sem COVID-19 e não realizavam intervenções instrumentais nas vias aéreas; e grupo de risco alto (GRA), profissionais que cuidavam de pacientes com COVID-19 e realizavam intervenções instrumentais com geração de aerossóis. Para estabelecer o desfecho, considerou-se a história de COVID-19 do profissional de saúde e a detecção de anticorpos IgG anti- SARS-CoV-2 por Elecsys® Anti-SARS-CoV-2. Resultados: A doença foi diagnosticada em 43,8% dos profissionais no GRB, 46,7% no GRM (razão de chances ajustada: 1,125; intervalo de confiança de 95% [IC95%]: 0,896-1,414; p = 0,311) e 48,6% no GRA (razão de chances: 1,214; IC95%: 0,964-1,530; p = 0,10). Conclusões: Pertencer ao GRM e ao GRA em função do nível de exposição a pacientes confirmados com COVID-19 no ambiente de trabalho não foi associado a um aumento da ocorrência da doença ou da soroconversão.

Performance Comparison of Object Detection Networks for Shrapnel Identification in Ultrasound Images.

Ultrasound imaging is a critical tool for triaging and diagnosing subjects but only if images can be properly interpreted. Unfortunately, in remote or military medicine situations, the expertise to interpret images can be lacking. Machine-learning image interpretation models that are explainable to the end user and deployable in real time with ultrasound equipment have the potential to solve this problem. We have previously shown how a YOLOv3 (You Only Look Once) object detection algorithm can be used for tracking shrapnel, artery, vein, and nerve fiber bundle features in a tissue phantom. However, real-time implementation of an object detection model requires optimizing model inference time. Here, we compare the performance of five different object detection deep-learning models with varying architectures and trainable parameters to determine which model is most suitable for this shrapnel-tracking ultrasound image application. We used a dataset of more than 16,000 ultrasound images from gelatin tissue phantoms containing artery, vein, nerve fiber, and shrapnel features for training and evaluating each model. Every object detection model surpassed 0.85 mean average precision except for the detection transformer model. Overall, the YOLOv7tiny model had the higher mean average precision and quickest inference time, making it the obvious model choice for this ultrasound imaging application. Other object detection models were overfitting the data as was determined by lower testing performance compared with higher training performance. In summary, the YOLOv7tiny object detection model had the best mean average precision and inference time and was selected as optimal for this application. Next steps will implement this object detection algorithm for real-time applications, an important next step in translating AI models for emergency and military medicine.

Treatment of intracardiac thrombi using ultra-slow low-dose thrombolytic therapy: A case report.

A 57-year-old male with previously known severe primary mitral regurgitation was admitted to the intensive care unit (ICU) due to massive venous thromboembolism, associated with right ventricular dysfunction and two large mobile right atrial thrombi. Due to deterioration in his clinical condition despite standard treatment with unfractionated heparin, it was decided to use an ultra-slow low-dose thrombolysis protocol, which consisted of a 24-hour infusion of 24 mg of alteplase at a rate of 1 mg per hour, without initial bolus. The treatment was continued for 48 consecutive hours, with clinical improvement and resolution of the intracardiac thrombi and no complications. One month after ICU admission, successful mitral valve repair surgery was conducted. This case demonstrates that ultra-slow low-dose thrombolysis is a valid bailout treatment option in patients with large intracardiac thrombi refractory to the standard approach.

Using Ultrasound Image Augmentation and Ensemble Predictions to Prevent Machine-Learning Model Overfitting.

Deep learning predictive models have the potential to simplify and automate medical imaging diagnostics by lowering the skill threshold for image interpretation. However, this requires predictive models that are generalized to handle subject variability as seen clinically. Here, we highlight methods to improve test accuracy of an image classifier model for shrapnel identification using tissue phantom image sets. Using a previously developed image classifier neural network-termed ShrapML-blind test accuracy was less than 70% and was variable depending on the training/test data setup, as determined by a leave one subject out (LOSO) holdout methodology. Introduction of affine transformations for image augmentation or MixUp methodologies to generate additional training sets improved model performance and overall accuracy improved to 75%. Further improvements were made by aggregating predictions across five LOSO holdouts. This was done by bagging confidences or predictions from all LOSOs or the top-3 LOSO confidence models for each image prediction. Top-3 LOSO confidence bagging performed best, with test accuracy improved to greater than 85% accuracy for two different blind tissue phantoms. This was confirmed by gradient-weighted class activation mapping to highlight that the image classifier was tracking shrapnel in the image sets. Overall, data augmentation and ensemble prediction approaches were suitable for creating more generalized predictive models for ultrasound image analysis, a critical step for real-time diagnostic deployment.

Mean-level correspondence and moment-to-moment synchrony in adolescent and parent affect: Exploring associations with adolescent age and internalizing and externalizing symptoms.

Interactions with parents are integral in shaping the development of children's emotional processes. Important aspects of these interactions are overall (mean level) affective experience and affective synchrony (linkages between parent and child affect across time). Respectively, mean-level affect and affective synchrony reflect aspects of the content and structure of dyadic interactions. Most research on parent-child affect during dyadic interactions has focused on infancy and early childhood; adolescence, however, is a key period for both normative emotional development and the emergence of emotional disorders. We examined affect in early to mid-adolescents (N = 55, Mage = 12.27) and their parents using a video-mediated recall task of 10-min conflict-topic discussions. Using multilevel modeling, we found evidence of significant level-2 effects (mean affect) and level-1 effects (affective synchrony) for parents and their adolescents. Level-2 and level-1 associations were differentially moderated by adolescent age and adolescent internalizing and externalizing symptoms. More specifically, parent-adolescent synchrony was stronger when adolescents were older and had more internalizing problems. Further, more positive adolescent mean affect was associated with more positive parent affect (and vice versa), but only for dyads with low adolescent externalizing problems. Results underscore the importance of additional research examining parent-child affect in adolescence.