Thoracic injuries caused by blunt trauma are commonly encountered emergencies in veterinary medicine. However, no literature exists comparing radiography to computed tomography (CT) in blunt thoracic trauma caused by motor vehicle accidents in canine patients. The aim of this prospective case series was to estimate the sensitivity (Se) and specificity (Sp) of thoracic radiography relative to CT for detecting lung contusions, pneumothorax, pleural effusion and rib fractures. The study further aimed to establish a severity scoring system for radiography and CT and to compare the findings between the two modalities. The hypothesis was that radiography would be less sensitive than CT at detecting these injuries and that radiography would underestimate the severity of lung contusions compared to CT. Fifty-nine patients met the inclusion criteria. Radiography underestimated the presence of lung contusions (Se = 69%, 95% Confidence interval (CI)) and overestimated the severity of the contusions relative to CT. There was also high interobserver variability in evaluating lung contusion severity (coefficient of variation = 91%). Both the three-view thoracic and horizontal beam radiography had poor sensitivities for the detection of pneumothorax (Se = 19% and 63% respectively) and pleural effusions (Se = 43% and 71% respectively). Similarly, the sensitivity (56%) of three-view thoracic radiographs for the detection of rib fractures was poor relative to CT. To conclude, three-view thoracic radiography had low sensitivity for pathology related to blunt thoracic trauma caused by motor vehicle accidents and CT could be considered as an additional diagnostic imaging modality in these patients. / Dissertation (MMedVet (Diagnostic Imaging))--University of Pretoria, 2019. / Companion Animal Clinical Studies / MMedVet (Diagnostic Imaging) / Unrestricted

京都大学 / 0048 / 新制・論文博士 / 博士(医学) / 乙第13158号 / 論医博第2145号 / 新制||医||1029(附属図書館) / (主査)教授 川上 浩司, 教授 戸井 雅和, 教授 松田 秀一 / 学位規則第4条第2項該当 / Doctor of Medical Science / Kyoto University / DFAM

The use of AI in the medical field is becoming more widespread, and research on its various applications is very popular. In biomedical image analysis, Convolutional Neural Networks (CNN), which are specialized in image processing, can analyze X-rays and detect signs of different diseases. However, to achieve that, CNNs require vast amounts of X-ray images with labels specifying the disease (labeled training data), which is not always available. One method to overcome this obstacle is the use of data augmentation. Data augmentation is manipulating images through flipping, rotating, or changing the saturation or brightness, among other methods. The purpose is to increase and diversify the training data to make the CNN model more robust. Our study aims to investigate the effects of different data augmentation techniques on the performance of a CNN model in detecting Pneumothorax. After fine-tuning our CNN model’s hyper-parameters, three data augmentation methods (color, geometric, and noise) and their combinations were applied to our model. We then tested and compared the effects of each data augmentation method on the accuracy of our model. Our study concluded that color augmentation performed the best compared to the other augmentation methods, while geometric augmentation had the worst performance. However, none of the augmentation methods significantly improved the original model’s performance, which can be attributed to the model’s configuration of hyper-parameters, leaving no room for improvement. / Användningen av AI inom det medicinska området blir mer utbredd och forskning om dess olika tillämpningar är mycket populär. Inom biomedicinsk bildanalys kan Convolutional Neural Networks (CNN), som är specialiserade på bildbehandling, analysera röntgenstrålar och upptäcka tecken på olika sjukdomar. Men för att uppnå det kräver CNN stora mängder röntgenbilder med etiketter som anger sjukdomen (märkta träningsdata), vilket inte alltid är tillgängligt. En metod för att övervinna detta hinder är användningen av dataförstärkning. Dataförstärkning är att manipulera bilder genom att bläddra, rotera eller ändra mättnad eller ljusstyrka, bland andra metoder. Syftet är att öka och diversifiera träningsdata för att göra CNN-modellen mer robust. Vår studie syftar till att undersöka effekterna av olika dataförstärkningstekniker på prestandan hos en CNN-modell vid detektering av pneumothorax. Efter att ha finjusterat vår CNN-modells hyperparametrar, tillämpades tre dataförstärkningsmetoder (färg, geometrisk och brus) och deras kombinationer på vår modell. Vi testade och jämförde sedan effekterna av varje dataförstärkningsmetod på noggrannheten i vår modell. Vår studie drog slutsatsen att färgförstärkning presterade bäst jämfört med andra förstärkningsmetoder, medan geometrisk förstärkning hade sämst prestanda. Ingen av förstärkningsmetoderna förbättrade dock den ursprungliga modellens prestanda avsevärt, vilket kan tillskrivas modellens konfiguration av hyperparametrar, vilket inte lämnar något utrymme för förbättringar.

Whether caused by trauma, internal diseases, or spontaneously; pneumothorax and hemothorax are potentially life threatening illnesses. They are currently primarily diagnosed with x-rays, CT scans, and ultrasound imaging. While these methods are generally reliable, they are not always available to the injured patient. Life threatening pneumothoraces, such as tension pneumothorax, must be treated quickly. When diagnosed correctly, pneumothorax and hemothorax can be quickly mended through insertion of chest tubes. X-rays, CT scans, and ultrasound imaging require large scale equipment and are not always dependable. There is a more reliable, portable, and faster result producing method to diagnose pneumothorax. Vibro-acoustic waves can be sent through the chest and the resulting wave can be measured. By analyzing attenuation characteristics determined by the geometry of the chest structures, it can be determined if the patient's pleural space is healthy, contains air (pneumothrax), or contains fluid (hemothorax).

Reproduction de : Thèse d'exercice : Médecine : Nancy 1 : 2004. / Titre provenant de l'écran-titre.

(Thesis: M.S.P.H.) PLEASE NOTE: Boston University Libraries did not receive an Authorization To Manage form for this thesis or dissertation. It is therefore not openly accessible, though it may be available by request. If you are the author or principal advisor of this work and would like to request open access for it, please contact us at open-help@bu.edu. Thank you. / INTRODUCTION: Although serious complications of invasive bedside procedures are rare they can be life threatening. Interest in identifying and preventing complications of bedside procedures has been heightened recently in light of such complications increasingly being linked to hospital reimbursement. In part I of this thesis we present the validation of ICD-9-CM codes for identification of the two most common complications associated with central venous catheterization. In part II we determine the impact of the development of a medical procedure service (MPS) on patient safety surrounding bedside procedures and resident education. METHODS: Part I: Cases of iatrogenic pneumothorax and central line associated blood stream infections (CLABSI) identified by ICD-9-CM codes in discharge data provided by Boston University Medical Center to the University HealthSystem Consortium were compared with those revealed by medical record abstraction of central venous catheters placed between 7/10-12/11. Part II: Retrospective cohort analysis of consecutive adults admitted to the internal medicine service who underwent a bedside medical procedure between 7/10-12/11 comparing characteristics and outcomes of procedures performed by the MPS versus primary services. RESULTS: Part I: The ICD-9-CM code for iatrogenic pneumothorax (512.1) had a sensitivity of 66.7%, specificity of 100%, positive predictive value of 100% and negative predictive value of 99.5%. The ICD-9-CM codes for CLABSI (999.31 and 999.32) had a sensitivity of 41.7%, specificity of 98.0%, positive predictive value of 20.0% and negative predictive value of 99.3%. Part II: We evaluated 1707 bedside procedures (548 by MPS, 1159 by primary services). While the MPS was more likely to successfully complete procedures (95.8% vs. 92.8%, p=0.02) and to use best practice safety process measures (95.4% vs. 51.0%, p<0.0001), the composite rate of major complications was similar (1.6% vs. 1.9%, p=0.71). CONCLUSIONS: Complications associated with invasive beside procedures are rare. The low sensitivity and variable positive predictive value of ICD-9-CM codes for detection of complications of central venous catheterization limits their use for internal quality improvement purposes. While use of a medical procedure service was associated with increased use of evidence based process measures, it did not significant affect the rate of major complications associated with bedside procedures. / 2031-01-01

Background: Tension pneumothorax is a condition with frequent fatal complications. This condition is caused by a disruption in the lung - that creates a one-way valve allowing air to accumulate in the pleural space. The fatal complication is the prevention of blood returning to the right side of the heart - due intrathoracic pressure compressing the right atrium. The patient can exhibit symptoms of dyspnea, tachypnea, tracheal deviation, jugular venous distention, subcutaneous emphysema, and shock that can lead to rapid deterioration and death. Case Report: We report a case of massive subcutaneous emphysema complicating tension pneumothorax management. The patient is a 20-year-old male who presented to the emergency department with chest trauma and was in extremis with diffuse severe subcutaneous emphysema. Due to the distorted anatomy, airway management and chest decompression were performed with nonstandard techniques/equipment resulting in rapid patient stabilization. After 4 days in the hospital, he was discharged home with no deficits. Discussion: Many providers do not have the proper equipment or training to treat patients in this extreme condition. CT images demonstrate the anatomical distortions in this case and the increase in size required for invasive life-saving devices. Images demonstrate where many commercial 14 gauge angiocaths and cricothyrotomy kits will not suffice (due to distortion in the anatomy), and these should not be relied on solely. Conclusions: While trauma carts frequently maintain (1.75 - 2 inch) 14 gauge angiocaths, they should also have military grade angiocaths that are 3.25” in length, which will work in most cases. Some, but not all, military-grade cricothyrotomy kits, or individually assembled kits, have 6.0 endotracheal tubes and come with a bougie and cricothyrotomy hook which would have been sufficient in this patient. Prehospital and hospital healthcare personnel should be prepared for similar patient encounters.

By taking advantage of the increasing amount of available electronic health data, applications of machine learning in the intensive care unit have the potential to improve medical diagnostics and risk stratification. This thesis proposes an approach for early onset prediction of pneumothorax with such technique, using time series data extracted from a clinical database. The prevalence of pneumothorax among patients is identified through ICD-9 codes, and labels for the onset are obtained by relying on proxies closely related to the condition. Both simple algorithms and deep learning networks are used in a sliding window-based prediction framework, and the importance of each feature is measured with weighted Shapley values. The results proved the feasibility of this approach using Long Short-Term Memory models, although the number of false positives is noticeably high. Mechanical ventilation was the most contributing feature for the outcome. In future work, the focus should be on addressing the large class imbalance that prevails, along with considering more well-founded methods of target labeling.

L’intoxication tabagique, source d’oxydants, est un facteur de risque important de développement de l’emphysème pulmonaire et du pneumothorax spontané primitif. Les macrophages alvéolaires contribuent pour une large part à l’inflammation pulmonaire au cours de ces pathologies en produisant des métalloprotéases et des espèces réactives de l’oxygène à l’origine du déséquilibre des balances protéase/anti-protéase et oxydant/antioxydant. L'hème oxygénase-1 (HO-1), exprimée principalement par les macrophages, est une enzyme clé des défenses anti-oxydantes pulmonaires. Nous avons initialement étudié l’expression et la localisation cellulaire de l’HO-1 et de ses régulateurs potentiels (Nrf2, Keap1, Bach1 et HIF-1a) dans les macrophages alvéolaires au cours de l’emphysème pulmonaire post-tabagique et du pneumothorax spontané primitif. Les voies de régulation de l’expression de ces protéines ont été analysées in vitro sur des macrophages dérivés de la lignée THP-1 exposés ou non au condensat de fumée de cigarette et à l’hypoxieréoxygénation visant à mimer une partie des effets de l’atélectasie-réexpansion observée lors de la prise en charge thérapeutique des pneumothorax récidivants. Les travaux présentés dans cette thèse nous ont permis de mettre en évidence une altération de l’expression de la voie Nrf2/Keap1-Bach1 associée à une diminution de l’expression des enzymes anti-oxydantes, dont l’HO-1, dans les macrophages alvéolaires au cours de l’emphysème pulmonaire sévère post-tabagique, malgré un stress oxydant important. In vitro, ces altérations pourraient être liées à une activation spécifique des MAPKinases ERK1/2 et JNK par le condensat de fumée de cigarette. Nous avons également montré que la stimulation du système de l’HO-1 était probablement orchestrée par la voie du facteur HIF-1a, et non par celle de Nrf2, dans les macrophages alvéolaires au cours du pneumothorax spontané primitif récidivant du sujet fumeur. Ces résultats pourraient contribuer à une meilleure connaissance de la physiopathologie de l’emphysème pulmonaire et permettre d’envisager de nouvelles approches thérapeutiques basées sur la préservation et/ou la restauration de l’équilibre Nrf2/Keap1-Bach1. Nos travaux suggèrent également que la physiopathologie du pneumothorax spontané primitif est différente chez les patients fumeurs et non fumeurs. Le pneumothorax du sujet fumeur est associé à un stress oxydant pulmonaire et à une induction de l’HO-1 probablement orchestrée par HIF-1a. Ces résultats, confirmés in vitro, mettent en évidence une interaction potentielle entre le stress oxydant et l’hypoxie-réoxygénation / Chronic cigarette smoking, a source of oxidants, is an important risk factor for lung emphysema and primary spontaneous pneumothorax development. Alveolar macrophages are mainly involved in lung inflammation observed in these pathologies through the production of metalloproteases and reactive oxygen species resulting to protease/anti-protease and oxidant/anti-oxidant imbalances. Heme oxygenase-1 (HO-1), mainly expressed in macrophages, is a key enzyme in pulmonary anti-oxidant defences. Therefore, the first aim of our studies was to investigate the expression and cellular localisation of HO-1 and its potential regulators (Nrf2, Keap1, Bach1 and HIF-1a) in alveolar macrophages from smoking related lung emphysema and primary spontaneous pneumothorax. Regulation pathways involved in expression of these proteins were assessed in vitro in macrophage cell line THP-1 exposed or not to cigarette smoke condensate and with or without hypoxia-reoxygenation mimicking parts of events induced by atelectasia-reexpansion during recurrent pneumothorax constitution and treatment. In these studies, we showed an altered expression of Nrf2/Keap1- Bach1 pathway associated with a reduced expression of anti-oxidants enzymes, like HO-1, in alveolar macrophages from smoking related lung emphysema patients, despite an important oxidative stress. These alterations might be related to cigarette smoke condensate activated ERK1/2 and JNK MAPKinases as observed in THP-1 cells. Furthermore, we showed that HO- 1 system induction was mediated by HIF-1a instead of Nrf2 pathway in alveolar macrophages from smoking related recurrent primary spontaneous pneumothorax. These findings may contribute to a better knowledge of the pathophysiology of lung emphysema and could provide new therapeutic approaches based on preservation and/or restoration of Nrf2/Keap1-Bach1 equilibrium. Our results also suggest that the pathophysiology of primary spontaneous pneumothorax could be different in smokers and non smokers. Spontaneous pneumothorax in smokers is associated with lung oxidative stress and the orchestrated induction of HO-1 probably via HIF-1a. These results provide a new link between oxidative stress and hypoxia/reoxygenation

Background: Global and regional transpulmonary pressure (PL) during one-lung ventilation (OLV) is poorly characterized. We hypothesized that global and regional PL and driving PL (ΔPL) increase during protective low tidal volume OLV compared to two-lung ventilation (TLV), and vary with body position. Methods: In sixteen anesthetized juvenile pigs, intra-pleural pressure sensors were placed in ventral, dorsal, and caudal zones of the left hemithorax by video-assisted thoracoscopy. A right thoracotomy was performed and lipopolysaccharide administered intravenously to mimic the inflammatory response due to thoracic surgery. Animals were ventilated in a volume-controlled mode with a tidal volume (VT) of 6 mL kg⁻¹ during TLV and of 5 mL kg⁻¹ during OLV and a positive end-expiratory pressure (PEEP) of 5 cmH₂O. Global and local transpulmonary pressures were calculated. Lung instability was defined as end-expiratory PL<2.9 cmH₂O according to previous investigations. Variables were acquired during TLV (TLVsupine), left lung ventilation in supine (OLVsupine), semilateral (OLVsemilateral), lateral (OLVlateral) and prone (OLVprone) positions randomized according to Latin-square sequence. Effects of position were tested using repeated measures ANOVA. Results: End-expiratory PL and ΔPL were higher during OLVsupine than TLVsupine. During OLV, regional end-inspiratory PL and ΔPL did not differ significantly among body positions. Yet, end-expiratory PL was lower in semilateral (ventral: 4.8 ± 2.9 cmH₂O; caudal: 3.1 ± 2.6 cmH₂O) and lateral (ventral: 1.9 ± 3.3 cmH₂O; caudal: 2.7 ± 1.7 cmH₂O) compared to supine (ventral: 4.8 ± 2.9 cmH₂O; caudal: 3.1 ± 2.6 cmH2O) and prone position (ventral: 1.7 ± 2.5 cmH₂O; caudal: 3.3 ± 1.6 cmH₂O), mainly in ventral (p ≤ 0.001) and caudal (p = 0.007) regions. Lung instability was detected more often in semilateral (26 out of 48 measurements; p = 0.012) and lateral (29 out of 48 measurements, p < 0.001) as compared to supine position (15 out of 48 measurements), and more often in lateral as compared to prone position (19 out of 48 measurements, p = 0.027). Conclusion: Compared to TLV, OLV increased lung stress. Body position did not affect stress of the ventilated lung during OLV, but lung stability was lowest in semilateral and lateral decubitus position.