Search Results (347)

Hyperechoic vertical artifacts are an essential feature of lung ultrasound (LUS) arising from various pathological states. Those that meet the criteria for B-lines have the most significant diagnostic value and should be differentiated from other hyperechoic vertical artifacts of unspecified clinical importance. Although numerous studies have assessed the impacts of transducer type on the appearance of B-lines in human medicine, comparative studies in veterinary medicine are limited and conflicting. This study compares three transducer types for the assessment of hyperechoic vertical artifacts in dogs. We hypothesize that there is high-level reviewer agreement in the assessment of HVA image quality and characteristics, and that the image quality/characteristics differ between the three transducers. Dogs (n = 8) with HVAs and sonographic absence of lung consolidations, pleural effusion, and/or pneumothorax were enrolled. Twenty-four cine-loops (5 s) containing HVAs were retrospectively and independently reviewed by two reviewers, who were blinded to the case details but not transducer type. The reviewers assessed the cine-loops for the following: whether HVAs meet the B-line criteria, ease of counting HVAs, and overall image quality. Paired cine-loops from the same patient using different transducers were then presented for HVA quality comparison. Inter-rater concordance was determined using the Kappa coefficient, Kendall’s tau, and Pearson correlation coefficient, while characteristics were compared using chi-square and Kruskal–Wallis tests (level of significance, α = 0.05). The overall concordance of image quality was good (Pearson’s coefficient = 0.82). The PA transducer scored lower in image quality (p < 0.001), HVA blending (p = 0.014), graininess (p < 0.001), and clarity of edges (p < 0.001) when compared with the microconvex and linear transducers, and the identification of B-line criteria differed between transducers (p = 0.024). Furthermore, the PA scored lowest in the comparison of paired cine-loops regarding the image and HVA quality (p < 0.001). Although more HVAs failed to reach the far field with the linear transducer (10/16, 62.5%) compared with the microconvex (8/16, 50%) and PA (3/16, 18.5%) transducers, the linear transducer scored higher than the microconvex and PA transducers regarding its ability to count B-lines (p < 0.001). This study demonstrates that the type of transducer significantly impacts the characteristics of HVAs, with the PA transducer producing lower-quality images compared with the microconvex and linear transducers. Full article

Background: The use of resuscitative endovascular balloon occlusion of the aorta (REBOA) for control of noncompressible torso hemorrhage remains controversial. We aimed to utilize a novel and transparent/interpretable artificial intelligence (AI) method called Optimal Policy Trees (OPTs) to improve the appropriate use and decrease the misuse of REBOA in hemodynamically unstable blunt trauma patients. Methods: We trained and then validated OPTs that “prescribe” REBOA in a 50:50 split on all hemorrhagic shock blunt trauma patients in the 2010–2019 ACS-TQIP database based on rates of survival. Hemorrhagic shock was defined as a systolic blood pressure ≤90 on arrival or a transfusion requirement of ≥4 units of blood in the first 4 h of presentation. The expected 24 h mortality rate following OPT prescription was compared to the observed 24 h mortality rate in patients who were or were not treated with REBOA. Results: Out of 4.5 million patients, 100,615 were included, and 803 underwent REBOA. REBOA patients had a higher rate of pelvic fracture, femur fracture, hemothorax, pneumothorax, and thoracic aorta injury (p < 0.001). The 24 h mortality rate for the REBOA vs. non-REBOA group was 47% vs. 21%, respectively (p < 0.001). OPTs resulted in an 18% reduction in 24 h mortality for REBOA and a 0.8% reduction in non-REBOA patients. We specifically divert the misuse of REBOA by recommending against REBOA in cases where it leads to worse outcomes. Conclusions: This proof-of-concept study shows that interpretable AI models can improve mortality in unstable blunt trauma patients by optimizing the use and decreasing the misuse of REBOA. To date, these models have been used to predict outcomes, but their groundbreaking use will be in prescribing interventions and changing outcomes. Full article

Introduction: Alpha-1 antitrypsin deficiency (AATD) is a genetic condition caused by SERPINA1 variants with variable severity. Current international guidelines do not recommend augmentation therapy for intermediate AATD; nevertheless, some patients show clinically severe phenotypes in real-world practice. We aimed to evaluate, in an exploratory manner, the potential effects of augmentation therapy on exacerbations, quality of life, and lung function in this subgroup. Methods: In this multicenter retrospective study, we included 27 heterozygous patients with intermediate AATD (serum AAT 50–110 mg/dL), Chronic Obstructive Pulmonary Disease (COPD), and/or emphysema. Clinical phenotypes included emphysema-predominant disease, COPD with frequent exacerbations, and overlap with bronchiectasis/asthma; HRCT patterns were recorded. We assessed the annual number of exacerbations (moderate: steroids/antibiotics; severe: hospitalization/including pneumothorax), St. George’s Respiratory Questionnaire (SGRQ), and lung function before and after 12 months of therapy. Results: Augmentation therapy was associated with a reduction in annual exacerbations from a median (IQR) of 2 (1.5–3) to 1 (0–1) (p < 0.0001) and an improvement in SGRQ total score (58.89 ± 16.83 to 48.34 ± 21.20; p = 0.0039). The mean SGRQ change exceeded the 4-point MCID for COPD. No significant changes were observed in spirometry or Diffusing Capacity of the Lung for Carbon Monoxide (DLCO). Conclusions: These exploratory findings suggest that augmentation therapy may reduce exacerbations and improve quality of life in selected patients with intermediate AATD and COPD/emphysema. Given the retrospective design, small sample, and lack of a control group, the results should be interpreted as hypothesis-generating and warrant confirmation in prospective studies. Full article

Introduction: Non-surgical biopsy is recommended for diagnosing solid pulmonary nodules measuring >8 mm when the probability of malignancy is low to moderate. However, currently available biopsy methods do not have a sufficient diagnostic yield for nodule size <20 mm. Previous work has shown that one-lung flooding (OLF) enables complete lung sonography and good demarcation of lung nodules. Therefore, here, we report the first experiences with ultrasound-guided transthoracic core needle biopsy (USgTTcNB) under OLF for the histological diagnosis of small pulmonary nodules. Methods: In two patients with small pulmonary nodules, a transbronchial/thoracic biopsy was not indicated due to the size and location of the nodules. Following nodule detection under OLF, the USgTTcNB was performed. The biopsy cylinder was immediately examined via the frozen section procedure. After liquid draining and re-ventilation, the patients were extubated in the operation room and monitored in the intermediate care unit. Results: In both patients, a histological diagnosis was achieved. In the case of malignancy, the patient underwent lobectomy during the same session. In the case of a benign diagnosis, a futile operation was avoided. In case two, a small apical pneumothorax occurred. The hemodynamic values during and after the intervention were in the normal range. Lung function on day 2 after the intervention increased compared with that before the intervention. Conclusions: USgTTcNB under OLF is feasible and enables a histological confirmation of small pulmonary nodules. Nevertheless, this new promising technique should be evaluated in a study with a larger cohort. Full article

Background/Objectives: Blunt trauma injury of the diaphragm is uncommon. Even after imaging examination, accurate diagnosis remains difficult. We sought to identify clinical factors that raise suspicion of such injuries, which can be applied during the initial evaluation of trauma patients. Methods: We retrospectively analyzed patients with blunt trauma who were diagnosed with diaphragmatic injury between January 2015 and July 2025. Demographic variables, clinical findings, operative records, and imaging findings were reviewed. Results: The most common mechanism of injury in patients with diaphragmatic injury was traffic accidents (64.2%), and 77.4% were identified as severe injuries with an Injury Severity Score (ISS) ≥ 16. Computed tomography (CT) scans of these patients frequently showed hemothorax, hemoperitoneum, and pneumothorax, but 49.1% of cases did not show diaphragmatic injury on preoperative imaging. In these patients, pneumothorax, lower rib fractures, and liver injury were more common. Notably, pneumothorax strongly suggested the possibility of diaphragmatic injury in patients where intrathoracic herniation was not clear. Conclusions: In patients with polytrauma and unstable vital signs, CT evaluation of torso injuries and careful interpretation are essential. Even when CT does not reveal diaphragmatic injury, suspicion should be elevated in cases with high ISS accompanied by pneumothorax, hemoperitoneum, hemothorax, lower rib fractures, or extremity injuries. If the injury mechanism further raises clinical suspicion, repeated physical examinations and imaging studies should be performed. When suspicion remains high, surgical intervention should be considered to confirm the diagnosis. Full article

Background: Achalasia is a primary esophageal motility disorder characterized by impaired relaxation of the lower esophageal sphincter (LES) and absent peristalsis, which increases the risk of aspiration during anesthesia. Peroral endoscopic myotomy (POEM) is a minimally invasive therapeutic approach requiring tailored anesthetic management. This study aimed to evaluate perioperative anesthesia management during POEM, focusing on ventilation parameters, intraoperative hemodynamics, laboratory changes, and the incidence and severity of postoperative complications. Methods: A retrospective analysis was conducted on 51 patients who underwent POEM between June 2016 and April 2025. Demographic features, anesthesia techniques, intraoperative physiologic parameters, hematologic profiles, and postoperative complications were evaluated. Standard preoperative fasting protocols were implemented. Rapid sequence induction (RSI) with propofol and rocuronium was followed by endotracheal intubation. Desflurane was used for maintenance anesthesia, with ventilation settings adjusted to limit end-tidal carbon dioxide (ETCO₂) elevation. Results: The median age of patients was 48 years, with a slight female (52.9%) predominance. Most patients were American Society of Anesthesiologists (ASA) II (64.7%) or ASA III (35.3%) scores and had comorbid hypertension (31.4%) or diabetes (11.8%). The median anesthesia duration was 180 min, and the peak inspiratory pressure remained stable at 25 mmHg. Oxygen saturation (SpO₂) improved during the procedure, while ETCO₂ increased from baseline to 49 mmHg by the end. Blood pressure declined transiently but recovered intraoperatively. Hematologic analysis showed significant increases in white blood cell (WBC) and neutrophils and mild decreases in hemoglobin, hematocrit, and platelets. Early postoperative complications included subcutaneous emphysema (19.6%), minor bleeding (9.8%), and pneumoperitoneum (7.84%). Two patients required tube thoracostomy due to pneumothorax, but no patient developed a complication requiring surgical exploration. During a median follow-up of 546 days, no mortality was reported. Long-term complications were infrequent, with gastroesophageal reflux disease (GERD) (3.92%) and esophagitis (1.96%) being the most notable. Conclusions: POEM can be performed safely with appropriate anesthetic management. Despite significant physiologic changes during carbon dioxide (CO₂) insufflation, no life-threatening complications occurred, and the majority of adverse events were minor and self-limiting. Close intraoperative monitoring and interdisciplinary coordination contribute to favorable perioperative outcomes. Full article

Background: Peripheral pulmonary lesions (PPLs) are increasingly detected due to widespread use of cross-sectional imaging and lung cancer screening. While cone-beam CT and robotic bronchoscopy have improved diagnostic accuracy, they remain resource-intensive and inaccessible in many settings. A novel technique employing simultaneous use of two bronchoscopes referred to as the “Kissing Probe Technique” was developed to provide real-time lesion localization and precise sampling using standard equipment. Methods: This retrospective, single-centre study included 43 patients with radiologically confirmed PPLs suspicious for malignancy. Under general anaesthesia with rigid bronchoscopy or continuous sedation with endotracheal intubation, two single-use bronchoscopes were introduced in parallel. The first (standard diameter) housed a radial EBUS probe for lesion localization, while the second (ultra-thin) guided a 1.1 mm cryoprobe to the lesion based on direct ultrasound and fluoroscopic confirmation. Cryobiopsies were performed once “kissing contact” between the radial probe and cryoprobe was established. Results: A total of 43 procedures were completed without major complications. The mean lesion size was 24.6 mm. Radial probe localization was successful in 86% of cases, and tool-contact confirmation was achieved in 35/43 patients (81%). The overall diagnostic yield was 83.7% (36/43). Bleeding occurred in 23% of cases and was managed conservatively without the need for escalation of care. No pneumothorax or equipment-related damage occurred. Conclusions: The “Kissing Probe Technique” is a safe and feasible approach for bronchoscopic sampling of PPLs. It offers a cost-effective alternative for real-time tool-in-lesion confirmation using widely available equipment. Further multicentre validation is warranted to confirm generalizability and cost-effectiveness. Full article

Background/Objectives: The incidence of dysrhythmia after blunt thoracic trauma varies in the literature from 8–75%, and the complication rate from these dysrhythmias is not well studied. The aims of this study are to (1) identify the incidence of dysrhythmia following blunt thoracic trauma, (2) identify risk factors associated with developing a dysrhythmia, and (3) identify the incidence of cardiac intervention after developing a dysrhythmia. We hypothesize that blunt thoracic trauma may result in post-injury dysrhythmias. Methods: This is a retrospective review of trauma patients ≥ 18 years with a blunt mechanism of injury at a Level 1 Trauma Center from 1/2010 to 3/2022. Patients were included if they had one of the following: rib fracture, sternal fracture, chest wall contusion, pneumothorax, hemothorax, chest pain, chest wall deformity, or chest wall crepitus. Patients were excluded if they had an Abbreviated Injury Scale Chest = 0 or if they had a pre-existing dysrhythmia. Univariate, multivariate, and multivariable statistical analyses were performed. Results: In total, 2943 patients met inclusion criteria. In total, 574 (19.5%) developed a dysrhythmia; 100 (17.4%) required a new antiarrhythmic at discharge. Patients who developed a dysrhythmia had a nearly two times greater likelihood of requiring cardiac intervention than those without a dysrhythmia (AOR: 1.79; p = 0.004). Additional risk factors for requiring cardiac intervention included Injury Severity Score (ISS) 16–25 and >25 (p < 0.001). Conclusions: The incidence of dysrhythmia after blunt thoracic injury is 19.5% at our level I trauma center. Based on our study, patients that were older, had an ISS > 25, had a history of previous cardiac disease, or required > 5 units of blood products were at an increased risk of developing a dysrhythmia following trauma. As such, future consideration should be given to extended guidelines in monitoring these vulnerable patients. Full article

Background: Spontaneous pneumothorax (SP) in adolescents has a high recurrence risk, particularly without surgical treatment. This study aimed to predict recurrence using machine learning (ML) algorithms applied to chest computed tomography (CT) and to visualize CT features associated with recurrence. Methods: We retrospectively reviewed 299 adolescents with conservatively managed SP from January 2018 to December 2022. Clinical risk factors were statistically analyzed. Chest CT images were evaluated using ML models, with performance assessed by AUC, accuracy, precision, recall, and F1 score. Gradient-weighted Class Activation Mapping (Grad-CAM) was used for visual interpretation. Results: Among 164 right-sided and 135 left-sided SP cases, recurrence occurred in 54 and 43 cases, respectively. Mean recurrence intervals were 10.5 ± 9.9 months (right) and 12.7 ± 9.1 months (left). Presence of blebs or bullae was significantly associated with recurrence (p < 0.001). Neural networks achieved the best performance (AUC: 0.970 right, 0.958 left). Grad-CAM confirmed the role of blebs/bullae and highlighted apical lung regions in recurrence, even in their absence. Conclusions: ML algorithms applied to chest CT demonstrate high accuracy in predicting SP recurrence in adolescents. Visual analyses support the clinical relevance of blebs/bullae and suggest a key role of apical lung regions in recurrence, even when blebs/bullae are absent. Full article

Background: Double-valve infective endocarditis (DVIE) accounts for 15–20% of all endocarditis and represents a challenge due to the increased incidence of embolic events and congestive heart failure compared to infective endocarditis (IE) affecting one valve. This study aims to evaluate patients’ characteristics, surgical procedures, complications, and mortality associated with DVIE in our tertiary hospital in Italy. The Endocarditis Registry STEADY includes patients admitted with IE from January 2009 to March 2024 (n = 398). Sixty-three of them (16%) had DVIE. Methods: We conducted a retrospective single-center observational study, analyzing demographic, clinical, and microbiological data in DVIE patients, comparing those treated surgically (surgical group, SG) with those treated medically (non-surgical group, NSG). Results: The groups were homogeneous in age, microbiological yields, type of valve involved, and risk factors for infective endocarditis. The surgical group presented significantly more cancer history, intracardiac complications, and new-onset arrhythmias compared to the non-surgical group. Median hospital stay was similar in both groups. In SG, the most common postoperative complication was new rhythm disorders; other complications such as cardiac tamponade, pericardial effusion, and pneumothorax were rare. In-hospital mortality was similar between groups; however, one-year survival was higher in the surgical group (72% vs. 54%, p = 0.031). In our series, 16 patients were over 75 years old (25%), and 7 of them (44%) underwent cardiac surgery. One-year survival in the surgical group was also higher in this subgroup. Conclusions: Surgical treatment, when indicated, may improve the prognosis of patients with DVIE, including elderly patients. Full article

The aim of this retrospective clinical study was to assess surgical duration and surgical and post-operative complications associated with Video-Assisted Thoracoscopic Surgery (VATS) and lung biopsy in horses with severe Equine Pasture Asthma (EPA) and paired control horses. Twelve horses (6 EPA-affected, 6 control) were sex, age and breed matched. Twenty-four thoracic surgeries were performed. Surgery of each matched pair (EPA-affected and healthy) was performed during asthma exacerbation (summer) and remission (winter). Surgical times were shorter with uncomplicated thoracoscopy (85 min) and significantly longer (p < 0.001) when intra-operative complications necessitated conversion to thoracotomy (156 min). The overall surgical time of EPA-affected horses during asthma exacerbation was significantly longer than control horses at any time point, predicted mean difference of 78 min (p < 0.05). When comparing EPA-affected horses to themselves during asthma exacerbation and remission, surgical times were significantly longer (p < 0.01) with a predicted mean difference of 98 min; this effect of seasonality did not occur amongst control horses. Intra-operative surgical complications (6/24) were evenly divided between EPA and control horses, however, only severe EPA horses in exacerbation were noted to have lung hyperinflation. Post-operative complications: fever, colic, hemothorax, pneumothorax, subcutaneous emphysema, surgical site infection, and/or laminitis occurred in 13/24 surgical procedures (54%). No fatalities resulted from these procedures. Full article

Background/Objectives: The extensive use of computed tomography (CT) has led to a significant increase in the detection of small and non-palpable pulmonary nodules, necessitating the use of invasive methods for definitive diagnosis. Video-assisted thoracoscopic surgery (VATS) has become the preferred procedure for nodule resections; however, intraoperative localization remains challenging, especially for deep or subsolid lesions. This study explores whether SPECT/CT improves the technical and clinical outcomes of radio-guided occult lesion localization (ROLL) before uniportal video-assisted thoracoscopic surgery (u-VATS). Methods: This is a retrospective study involving consecutive patients referred for the resection of pulmonary nodules who underwent CT-guided ROLL followed by u-VATS between September 2017 and December 2024. From January 2023, SPECT/CT was systematically added after planar imaging. The cohort was divided into a planar group and a planar + SPECT/CT group. The inclusion criteria involved nodules sized ≤ 2 cm, with ground glass or solid characteristics, located at a depth of <6 cm from the pleural surface. ^99mTc-MAA injected activity, timing, the classification of planar and SPECT/CT image findings (focal uptake, multisite with focal uptake, multisite without focal uptake), spillage, and post-procedure complications were evaluated. Statistical analysis was performed, with continuous data expressed as the median and categorical data as the number. Comparisons were made using chi-square tests for categorical variables and the Mann–Whitney U test for procedural duration. Cohen’s kappa coefficient was calculated to assess agreement between imaging modalities. Results: In total, 125 patients were selected for CT-guided radiotracer injection followed by uniportal-VATS. The planar group and planar + SPECT/CT group comprised 60 and 65 patients, respectively. Focal uptake was detected in 68 (54%), multisite with focal uptake in 46 (36.8%), and multisite without focal uptake in 11 patients (8.8%). In comparative analyses between planar and SPECT/CT imaging in 65 patients, 91% exhibited focal uptake, revealing significant differences in classification for 40% of the patients. SPECT/CT corrected the classification of 23 patients initially categorized as multisite with focal uptake to focal uptake, improving localization accuracy. The mean procedure duration was 39 min with SPECT/CT. Pneumothorax was more frequently detected with SPECT/CT (43% vs. 1.6%). The intraoperative localization success rate was 96%. Conclusions: SPECT/CT imaging in the ROLL procedure for detecting pulmonary nodules before u-VATs demonstrates a significant advantage in reclassifying radiotracer positioning compared to planar imaging. Considering its limited impact on surgical success rates and additional procedural time, SPECT/CT should be reserved for technically challenging cases. Larger sample sizes, multicentric and prospective randomized studies, and formal cost–utility analyses are warranted. Full article

Background and Objectives: Neonatal respiratory distress syndrome (NRDS), resulting from a deficiency of pulmonary surfactant (PS), can cause alveoli to collapse. Glucocorticoids reduce inflammation and are effective in reducing pulmonary swelling. This study aims to assess the effectiveness of the combination of PS and budesonide in the management of NRDS. Materials and Methods: Publications between 21 May and 24 November were screened through PubMed, Cochrane and Embase. Data analysis was performed on RevMan 5.3 software. Subgroup analysis was performed to evaluate the routes of administrations. Results: The use of budesonide along with pulmonary surfactant for treating NRDS revealed the following results: (1) a reduced duration of invasive mechanical ventilation (standardized mean difference (SMD) = −1.06, 95% confidence interval (CI) = −1.55 to −0.56, p < 0.0001); (2) reduced rate of bronchopulmonary dysplasia (BPD) occurrence (relative risk (RR) = 0.72, 95% CI = 0.60 to 0.86, p = 0.0003); (3) reduced duration for hospital admittance (SMD = −0.38, 95% CI = −0.64 to −0.11, p = 0.005). The occurrence of complications, i.e., sepsis, pneumothorax, retinopathy of prematurity (ROP), necrotizing enterocolitis (NEC), rate of mortality, hyperglycemia and intraventricular hemorrhage (IVH), was not significantly different among the intervention and comparison group except for patent ductus arteriosus (PDA) and pulmonary hemorrhage, with their incidence being higher in the control group (p = 0.002 and p = 0.05, respectively). Conclusions: The combination of pulmonary surfactant and budesonide decreases the occurrence of BPD, duration of mechanical ventilation, length of hospital stay and risk of pulmonary hemorrhage and PDA. It does not increase the risk of complications and death and is clinically safe. Full article

Background/Objectives: The correlation between drugs and interstitial lung disease (ILD) is reported, but the presence of the substances of abuse in the lung as a cause of disease has never been proved. In this observational study, our aim was to evaluate a possible correlation between ILD radiological findings and cannabinoids presence in broncho-alveolar lavage (BAL) or in resected lung tissue in patients with a history of cannabis smoke. Methods: Data of patients with ILD chest CT scan findings and history of drug use, submitted to BAL (Group 1), or to lung apex removal for pneumothorax (Group 2), were retrospectively collected. In both groups, drug presence was investigated. A subgroup of Group 1 was checked for the concomitant presence in blood. Fisher’s test was used to study the association between the detection of the drug and ILD. Results: In Group 1, cannabinoids were present in 12/26 (46.2%) BAL samples. ILD emerged on chest CT in 75% of the cannabinoid-positive and in 20% of the cannabinoid-negative BAL samples (p = 0.0299). In the subgroup, the patients who tested positive for cannabinoids/cocaine on BAL were 55.6%; 0% were positive only on blood (p = 0.0294). In Group 2, cannabinoids were present in 10/15 (66.7%) specimens. ILD was evident, respectively, in 40% and in 0% of the patients with cannabinoid-positive and cannabinoid-negative surgical specimens (p = 0.2308). Conclusions: The prevalence of ILD in patients with cannabinoid-positive BAL and in those with cannabinoid-positive surgical specimens suggests that ILD could be caused by cannabis smoke. The non-concomitant presence of substances in BAL and in blood advocates the diagnostic usefulness of searching for the drug in the target organ. Full article

Background: Clinical imaging is an important part of health care providing physicians with great assistance in patients treatment. In fact, segmentation and grading of tumors can help doctors assess the severity of the cancer at an early stage and increase the chances of cure. Despite that Deep Learning for cancer diagnosis has achieved clinically acceptable accuracy, there still remains challenging tasks, especially in the context of insufficient labeled data and the subsequent need for expensive computational ressources. Objective: This paper presents a lightweight classification and segmentation deep learning model to assist in the identification of cancerous tumors with high accuracy despite the scarcity of medical data. Methods: We propose a multi-task architecture for classification and segmentation of cancerous tumors in the Brain, Skin, Prostate and lungs. The model is based on the UNet architecture with different pre-trained deep learning models (VGG 16 and MobileNetv2) as a backbone. The multi-task model is validated on relatively small datasets (slightly exceed 1200 images) that are diverse in terms of modalities (IRM, X-Ray, Dermoscopic and Digital Histopathology), number of classes, shapes, and sizes of cancer pathologies using the accuracy and dice coefficient as statistical metrics. Results: Experiments show that the multi-task approach improve the learning efficiency and the prediction accuracy for the segmentation and classification tasks, compared to training the individual models separately. The multi-task architecture reached a classification accuracy of 86%, 90%, 88%, and 87% respectively for Skin Lesion, Brain Tumor, Prostate Cancer and Pneumothorax. For the segmentation tasks we were able to achieve high precisions respectively 95%, 98% for the Skin Lesion and Brain Tumor segmentation and a 99% precise segmentation for both Prostate cancer and Pneumothorax. Proving that the multi-task solution is more efficient than single-task networks. Full article