Search

Simple Summary: Breast cancer affects millions of women worldwide. We aim to provide radiologists with automatic support for mammography review. We propose deploying deep learning models with Multiple Instance Learning algorithms for breast cancer diagnosis (cancer versus non-cancer classification) based on digital mammography images, taking advantage of data annotated at an image level without indicating where the lesion is. We employed algorithms to analyse original, high-resolution images with minimal reduction in size. We used graphical suggestions of attentional maps to verify the correctness of the algorithm and to indicate the areas of the images where cancer lesions are. Finally, we performed comparative and validation studies on external datasets differing in the number of images, pixel intensity levels, and subtypes of existing lesions, showing high accuracy and potential for generalisability of the algorithms. Breast cancer is the most frequent female cancer, with a considerable disease burden and high mortality. Early diagnosis with screening mammography might be facilitated by automated systems supported by deep learning artificial intelligence. We propose a model based on a weakly supervised Clustering-constrained Attention Multiple Instance Learning (CLAM) classifier able to train under data scarcity effectively. We used a private dataset with 1174 non-cancer and 794 cancer images labelled at the image level with pathological ground truth confirmation. We used feature extractors (ResNet-18, ResNet-34, ResNet-50 and EfficientNet-B0) pre-trained on ImageNet. The best results were achieved with multimodal-view classification using both CC and MLO images simultaneously, resized by half, with a patch size of 224 px and an overlap of 0.25. It resulted in AUC-ROC = 0.896 ± 0.017, F1-score 81.8 ± 3.2, accuracy 81.6 ± 3.2, precision 82.4 ± 3.3, and recall 81.6 ± 3.2. Evaluation with the Chinese Mammography Database, with 5-fold cross-validation, patient-wise breakdowns, and transfer learning, resulted in AUC-ROC 0.848 ± 0.015, F1-score 78.6 ± 2.0, accuracy 78.4 ± 1.9, precision 78.8 ± 2.0, and recall 78.4 ± 1.9. The CLAM algorithm's attentional maps indicate the features most relevant to the algorithm in the images. Our approach was more effective than in many other studies, allowing for some explainability and identifying erroneous predictions based on the wrong premises. [ABSTRACT FROM AUTHOR]

Introduction: Absent today is a simple numerical system of outcomes assessment that recognizes that bariatric surgery is metabolic surgery and incorporates weight loss, hypertension control, and type 2 diabetes control. Aim: To introduce a simple, new Numerical Scale to Assess the Outcomes of Metabolic Surgery (NOMS). Material and methods: For the stratification of weight outcomes, we used the percentage excess weight loss (%EWL); for hypertension, the systolic blood pressure (SBP) and diastolic blood pressure (DBP) combined with medication usage; and for type 2 diabetes, the hemoglobin A1c (HbA1c) value combined with medication usage. Results: Utilizing the guidelines of the American Diabetes Association, the Working Group of the European Society of Hypertension, the European Society of Cardiology, and the American College of Cardiology/American Heart Association, we propose for %EWL: W1 ≥ 50, W2 > 25 and < 50, and W3 ≤ 25; for hypertension H1 SBP/DPB < 140/90 mm Hg on no medication, H2 SBP/DBP ≥ 140/90 mm Hg with improvement of SBP or possible reduction of antihypertensive medication, and H3 no change or SBP higher than before surgery; for diabetes mellitus D1 HbA1c ≤ 7% and no medication, D2 HbA1c > 7% with a decrease of the HbA1c level or possible reduction of medication, D3 no change in HbA1c or HbA1c higher than before surgery. Designations of H0 and D0 are given if hypertension or diabetes was not present before surgery. Patient examples for numerical scores are provided. Conclusions: The introduction of our numerical scale (NOMS) can be of benefit in metabolic/bariatric outcomes assessment; communications among metabolic/bariatric surgery centers, physicians, and patients; and for more precise reporting in the evidence-based literature. [ABSTRACT FROM AUTHOR]

We describe an extracorporeal staple technique used to treat severe colostomy stenosis under analgo-sedation, thus avoiding relaparotomy. The surgery is performed under short-term sedation. The orifice of the stoma is widened and overgrowing skin is excised. The volume and diameter of the stoma are assessed. The anvil of a circular stapler device is inserted into the lumen of the colostomy. First bowel layers and then skin are closed with purse-string sutures. One firing of the stapler is used to reshape the stoma. The procedure takes around 20-30 min. One circular stapler is used. The patient can be discharged the same day or a day after surgery. No complications were noted in operated patients. At 6- and 12-month follow-ups, a slight narrowing of the colostomy was visible, but no recurrence of the stricture was noted. The described technique is an interesting, easy and safe alternative to previous methods of treatment for stenosed end-colostomy. Importantly, it is an extra-abdominal procedure and may be offered to patients with a history of multiple abdominal operations or with serious coexisting medical conditions in the one-day surgery setting. [ABSTRACT FROM AUTHOR]

Introduction: Laparoscopic sleeve gastrectomy (LSG) and laparoscopic adjustable gastric banding (LAGB) are acceptable options for primary bariatric procedures in patients with body mass index (BMI) 35-55 kg/m². Aim: The aim of this study is to compare the effects of these two bariatric procedures 6,12 and 24 months after surgery. Material and methods: Two hundred and two patients were included 72 LSG and 130 LAGB patients. The average age was 38.8 ± 11.9 and 39.4 ± 10.4 years in LSG and LAGB groups, with initial BMI of 44.1 kg/m² and 45.2 kg/m², p = NS. Results: The mean percentage of excess weight loss (%EWL) at 6 months for LSG vs. LAGB was 36.3% vs. 30.1% (p = 0.01) and at 12 months was 43.8% vs. 34.6% (p = 0.005). The greatest difference in the mean %EWL at 12 months was observed in patients with initial BMI of 40-49.9 kg/m² in favor of LSG (47.5% vs. 35.6%; p = 0.01). Two years after surgery there was no advantage of LSG and in the subgroup of patients with BMI 50-55 kg/m² there was a trend in favor of LAGB (57.2% vs. 30%; p = 0.07). The multiple regression model of independent variables (age, gender, initial BMI and the presence of comorbidities) proved insignificant in prediction of the best outcome in means of %EWL for either operative modality. None of these factors in the logistic regression model could determine the type of surgery that should be used in particular patients. Conclusions: During the first 2years after surgery, the best results were obtained in women with lower BMI undergoing LSG surgery. The LSG provides greater %EWL after a shorter period of time though the difference decreases in time. [ABSTRACT FROM AUTHOR]

Patients with gastric tumors usually present with symptoms of discomfort or pain in the epigastrium, regurgitations, nausea, vomiting or melena. Treatment options include open and laparoscopic total or partial gastrectomy and recently endoscopic mucosal resection. A case of successful endoscopic submucosal dissection is described with the unusual pathological finding of heterotopic pancreatic tissue forming a gastric tumor. The 67-year-old male patient was operated on due to the initial diagnosis of gastro-intestinal stromal tumor of the gastric trunk. Two intra-operative biopsies were negative for cancer cells. Submucosal endoscopic dissection was performed with IT and Hook knives (Olympus). A literature review was performed. The operative time was 180 min with hospital stay of 6 days. During the injection of the carmine dye and the air insufflation pneumoperitoneum occurred and remained clinically silent during the observation period. The pathology result showed a heterotopic pancreatic tissue type 2 according to Heinrich's classification with microfoci of intestinal metaplasia. Preoperative diagnostics of gastric masses might be misleading and such tumors not necessarily should be excised. There are several surgical options with endoscopic submucosal dissection being probably the safest one and a non-disabling approach. Patients tolerate that kind of surgery well with good postoperative functional outcomes. [ABSTRACT FROM AUTHOR]

Artificial intelligence (AI) is revolutionizing the field of medical imaging, holding the potential to shift medicine from a reactive "sick-care" approach to a proactive focus on healthcare and prevention. The successful development of AI in this domain relies on access to large, comprehensive, and standardized real-world datasets that accurately represent diverse populations and diseases. However, images and data are sensitive, and as such, before using them in any way the data needs to be modified to protect the privacy of the patients. This paper explores the approaches in the domain of five EU projects working on the creation of ethically compliant and GDPR-regulated European medical imaging platforms, focused on cancer-related data. It presents the individual approaches to the de-identification of imaging data, and describes the problems and the solutions adopted in each case. Further, lessons learned are provided, enabling future projects to optimally handle the problem of data de-identification. Critical relevance statement: This paper presents key approaches from five flagship EU projects for the de-identification of imaging and clinical data offering valuable insights and guidelines in the domain. Key Points: ΑΙ models for health imaging require access to large amounts of data. Access to large imaging datasets requires an appropriate de-identification process. This paper provides de-identification guidelines from the AI for health imaging (AI4HI) projects. [ABSTRACT FROM AUTHOR]

Purpose: This review aims to explore and summarise current knowledge of indocyanine green (ICG) fluorescence application in metabolic and bariatric surgery (MBS) and assess its potential in improving patient safety. Methods: The review was prepared according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) recommendations. Evidence from PubMed, ScienceDirect and Ovid MEDLINE databases was independently screened in October 2022. The primary information and outcomes were the type of fluorescence application with potential patient benefit and the complication rate. The secondary outcomes consisted of the kind of dye, the application protocol, and the equipment used. Results: Thirteen publications were included, with six prospective observational studies, five case reports and two retrospective cohort studies and involved a total of 424 patients. The publications were categorized into four groups based on the method of fluorescence application. Sixty-six percent of the cases used fluorescence for LSG, 32.3% used it for RYGB, 1.2% for revisional surgery, 0.2% during removal of an adjustable gastric band, and 0.2% for LSG combined with Rossetti fundoplication. ICG was used on its own in the majority of studies, although in one case, it was used in combination with methylene blue. ICG administration protocols varied significantly. Complications occurred in three patients (0.71%): leaks were diagnosed in two cases, and one patient required a blood transfusion. Conclusion: The most popular type of use is intraoperative assessment of the blood supply. ICG application may reduce the risk of complications, with potential benefits in detecting ischemia and leaks, searching for bleeding in areas inaccessible to endoscopy, and non-invasive hepatopathy evaluation. [ABSTRACT FROM AUTHOR]

Introduction: Pleural empyema is the most serious, life-threatening postoperative complication of pneumonectomy, observed after 1-12% of all pneumonectomies, with bronchopleural fistula being its main cause. Aim: The aim of this publication is to present early outcomes of minimally invasive surgical management of pleural empyema. Patients were subjected to a single, complex procedure, consisting of the laparoscopic mobilization of the greater omentum and its transposition via the diaphragm into the pleural cavity to fill in the empyema cavity with the consecutive pleuro-cutaneous fistuloplasty (thoracoplasty). Material and methods: Between May 2011 and April 2013, 8 patients were qualified to undergo the procedure. The mean age was 61 years (range: 46-77 years). Presence of bronchopleural fistula was confirmed in 3 cases. The median time of treatment with thoracostomy was 14.5 months. Results: The mean operative time was 125 min. The mean duration of post-operative hospital stay was 13.5 days (range: 7-31 days). In 6 patients (75%) the objective of permanent resolution of pleural empyema was achieved. In total, 4 patients had complications: pleural empyema recurrence (2 patients), splenic injury, hiatal hernia, gastrointestinal bleed. Two patients with empyema recurrence had Staphylococcus aureus infections prior to surgery. They were successfully managed both with prolonged thoracic drainage and antibiotics. Conclusions: Use of the greater omentum that was laparoscopically mobilized and transpositioned into the pleural cavity allows simultaneous management of the pleural empyema cavity and thoracostomy. The procedure is safe, with few direct complications. It is well tolerated and has at least a satisfactory cosmetic effect. The minimally invasive approach allows faster recovery and return to daily activities in comparison to the fully open technique. [ABSTRACT FROM AUTHOR]

Introduction: Hemorrhoidal disease is one of the commonest anorectal disorders worldwide. Stapled hemorrhoidopexy (SH) is a treatment modality associated with low postoperative pain and early mobilization. Aim: To assess long-term outcomes after SH. Material and methods: All 326 patients who underwent SH in 1999--2003 were invited by mail to participate. For each patient we analyzed their medical records, and conducted a questionnaire survey and a digital rectal examination. Results: Only 91 patients attended the final examination and the mean ± SD follow-up time was 8.7 ±1.2 years. Recurrences were diagnosed in one third of the 91 subjects. There were correlations between recurrences and: the duration of disease (p = 0.047); female gender (p = 0.037); and childbirth (vaginal delivery) (p = 0.026). Sixty-seven patients (73.6%) were satisfied with the outcomes. In the group of dissatisfied patients symptoms such as pain (p = 0.0001), burning (p = 0.0002) and itching (p = 0.014) were most common. Long-term outcomes were good with 75% and 88% reductions in pain sensation and severe and moderate hemorrhoidal bleeding. Pruritus, burning and discomfort resolved in more than 50% of patients. Flatus incontinence, fecal incontinence, or soiling occurred in 21%, 11%, and 32% of patients. Conclusions: Long-term results of stapled hemorrhoidopexy are satisfactory in most patients. The 36% recurrence rate correlates with the degree of hemorrhoidal prolapse before the operation, duration of the disease, female gender, and previous vaginal delivery. [ABSTRACT FROM AUTHOR]

To achieve clinical impact in daily oncological practice, emerging AI-based cancer imaging research needs to have clearly defined medical focus, AI methods, and outcomes to be estimated. AI-supported cancer imaging should predict major relevant clinical endpoints, aiming to extract associations and draw inferences in a fair, robust, and trustworthy way. AI-assisted solutions as medical devices, developed using multicenter heterogeneous datasets, should be targeted to have an impact on the clinical care pathway. When designing an AI-based research study in oncologic imaging, ensuring clinical impact in AI solutions requires careful consideration of key aspects, including target population selection, sample size definition, standards, and common data elements utilization, balanced dataset splitting, appropriate validation methodology, adequate ground truth, and careful selection of clinical endpoints. Endpoints may be pathology hallmarks, disease behavior, treatment response, or patient prognosis. Ensuring ethical, safety, and privacy considerations are also mandatory before clinical validation is performed. The Artificial Intelligence for Health Imaging (AI4HI) Clinical Working Group has discussed and present in this paper some indicative Machine Learning (ML) enabled decision-support solutions currently under research in the AI4HI projects, as well as the main considerations and requirements that AI solutions should have from a clinical perspective, which can be adopted into clinical practice. If effectively designed, implemented, and validated, cancer imaging AI-supported tools will have the potential to revolutionize the field of precision medicine in oncology. [ABSTRACT FROM AUTHOR]

Introduction: According to meta-analyses laparoscopic appendectomy is associated with many benefits. However, in comparison to open surgery an increased rate of intraabdominal abscesses (IAA) has been reported. Identification of predictive factors for this complication may help to identify patients with higher risk of IAA. Aim: To identify potential risk factors for intraabdominal abscess after laparoscopic appendectomy (LA). Material and methods: Eighteen surgical units in Poland and Germany submitted data of patients undergoing LA to the online web-based database created by the Polish Videosurgery Society of the Association of Polish Surgeons. It comprised 31 elements related to the pre-, intra- and postoperative period. Surgical outcomes were compared among the groups according to occurrence of IAA. Univariate and multivariate logistic regression models were used to identify potential risk factors for IAA. Results: 4618 patients were included in the analysis. IAA were found in 51 (1.10%) cases. Although several risk factors were found in univariate analysis, in the multivariate model, only the presence of complicated appendicitis was statistically significant (OR = 2.98, 95% CI: 1.11–8.04). Moreover, IAA has a significant influence on postoperative reintervention rate (OR = 126.95, 95% CI: 67.98–237.06), prolonged length of stay > 8 days (OR = 41.32, 95% CI: 22.86–74.72) and readmission rate (OR = 33.89, 95% CI: 18.60–34.73). Conclusions: Intraabdominal abscesses occurs relatively rarely after LA. It is strongly associated with complicated appendicitis. Occurrence of this complication has a great influence on the postoperative period and due to the nature of its treatment is associated with the need for reintervention, prolonged length of stay and by extension possible readmission. [ABSTRACT FROM AUTHOR]

More than 1 million inguinal hernia repairs are performed in Europe and the US annually. Although antibiotic prophylaxis is not required in clean, elective procedures, the routine use of implants (90% of inguinal hernia repairs are performed with mesh) makes the topic controversial. The European Hernia Society does not recommend routine antibiotic prophylaxis for elective inguinal hernia repairs. However, the latest randomized controlled trial, published by Mazaki et al., indicates that the use of prophylaxis is effective for the prevention of surgical site infection. Unnecessary prophylaxis contributes to the development of bacterial resistance and significantly increases healthcare costs. This review documents clinical trials on inguinal hernia repairs with mesh and summarizes the current knowledge. It also tries to solve certain problems, namely: what constitutes a real risk factor, late-onset infection, and how the "surgical environment" impacts on the need to use antibiotic prophylaxis. [ABSTRACT FROM AUTHOR]

Purpose: Preoperative nutrition is beneficial for malnourished cancer patients. Yet, there is little evidence whether or not it should be given to nonmalnourished patients. The aim of this study was to assess the need to introduce preoperative nutritional support in patients without malnutrition at qualification for surgery. Methods: This was a prospective, two-arm, randomized, controlled, open-label study. Patients in interventional group received nutritional supplementation for 14 days before surgery, while control group kept on to their everyday diet. Each patient's nutritional status was assessed twice-at qualification (weight loss in 6 months, laboratory parameters: albumin, total protein, transferrin, and total lymphocyte count) and 1 day before surgery (change in body weight and laboratory parameters). After surgery, all patients were followed up for 30 days for postoperative complications. Results: Fifty-four patients in interventional and 48 in control group were analyzed. In postoperative period, patients in control group suffered from significantly higher ( p < 0.001) number of serious complications compared with patients receiving nutritional supplementation. Moreover, levels of all laboratory parameters declined significantly ( p < 0.001) in these patients, while in interventional arm were stable (albumin and total protein) or raised (transferrin and total lymphocyte count). Conclusions: Preoperative nutritional support should be introduced for nonmalnourished patients as it helps to maintain proper nutritional status and reduce number and severity of postoperative complications compared with patients without such support. [ABSTRACT FROM AUTHOR]

Purpose: Preoperative nutrition is beneficial for malnourished cancer patients. Yet, there is little evidence whether or not it should be given to nonmalnourished patients. The aim of this study was to assess the need to introduce preoperative nutritional support in patients without malnutrition at qualification for surgery.Methods: This was a prospective, two-arm, randomized, controlled, open-label study. Patients in interventional group received nutritional supplementation for 14 days before surgery, while control group kept on to their everyday diet. Each patient's nutritional status was assessed twice--at qualification (weight loss in 6 months, laboratory parameters: albumin, total protein, transferrin, and total lymphocyte count) and 1 day before surgery (change in body weight and laboratory parameters). After surgery, all patients were followed up for 30 days for postoperative complications.Results: Fifty-four patients in interventional and 48 in control group were analyzed. In postoperative period, patients in control group suffered from significantly higher (p < 0.001) number of serious complications compared with patients receiving nutritional supplementation. Moreover, levels of all laboratory parameters declined significantly (p < 0.001) in these patients, while in interventional arm were stable (albumin and total protein) or raised (transferrin and total lymphocyte count).Conclusions: Preoperative nutritional support should be introduced for nonmalnourished patients as it helps to maintain proper nutritional status and reduce number and severity of postoperative complications compared with patients without such support. [ABSTRACT FROM AUTHOR]

Breast cancer has high metastatic potential with distant metastases involving mainly lungs, liver and bones. Less frequently it gives distant spread to other organs. Herein we would like to present a very rare case of an acute cholecystitis which turned out to be a metastatic breast cancer in previously healthy woman. A female patient, 64-years old, presented to the emergency department with symptoms of biliary colic and acute abdomen. During the emergency cholecystectomy, we diagnosed the gallbladder empyema with thickened wall. There were also multiple metastatic nodules in the peritoneal cavity and an excessive amount of free fluid. The emergency physicians diagnosing female patient with the acute abdominal symptoms and a breast cancer history might suspect malignant spread into abdominal organs including gallbladder. On the other hand, acute cholecystitis symptoms might be the first symptoms of metastatic process in the gallbladder from the unknown primary source, which may be breast. [ABSTRACT FROM AUTHOR]

Introduction: With the introduction of new surgical equipment, there is always the need for new, more advanced training. The authors try to answer whether the use of the newest generation tools has an impact on achieving better results in single incision laparoscopic surgery (SILS) technique during the exercises in the surgical skills laboratory. Material and methods: There were 51 participants in the study: 44 'novices' and 7 'experts'. All subjects performed the 'advanced grasping' exercise according to the FLS programme manual using four types of laparoscopic approach including two SILS ports and SILS-dedicated instruments. The outcome measures involved task completion time and the number of errors. Results: Tasks using straight laparoscopic instruments set together with classic three-port access as well as SILS access ports were finished significantly faster when compared with SILS-dedicated instruments (p < 0.05). There were no significant differences in performance times between the two setups with straight instruments (p < 0.05) and both setups with SILS-dedicated instruments, irrespective of the use of curved or dynamic articulated tools. Students with no previous laparoscopic experience had significantly worse task completion times in all tasks in comparison to students with laparoscopic laboratory training and the 'experts' group. Conclusions: The use of the straight instruments in the SILS technique remain similar to its performance in full triangulation. SILS-dedicated instruments paradoxically increase the task completion time irrespective of possessed skills. The study showed the necessity of a SILS-dedicated tools training programme. [ABSTRACT FROM AUTHOR]