Cardiovascular/Stroke Risk Stratification in Diabetic Foot Infection Patients Using Deep Learning-Based Artificial Intelligence: An Investigative Study

A diabetic foot infection (DFI) is among the most serious, incurable, and costly to treat conditions. The presence of a DFI renders machine learning (ML) systems extremely nonlinear, posing difficulties in CVD/stroke risk stratification. In addition, there is a limited number of well-explained ML paradigms due to comorbidity, sample size limits, and weak scientific and clinical validation methodologies. Deep neural networks (DNN) are potent machines for learning that generalize nonlinear situations. The objective of this article is to propose a novel investigation of deep learning (DL) solutions for predicting CVD/stroke risk in DFI patients. The Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) search strategy was used for the selection of 207 studies. We hypothesize that a DFI is responsible for increased morbidity and mortality due to the worsening of atherosclerotic disease and affecting coronary artery disease (CAD). Since surrogate biomarkers for CAD, such as carotid artery disease, can be used for monitoring CVD, we can thus use a DL-based model, namely, Long Short-Term Memory (LSTM) and Recurrent Neural Networks (RNN) for CVD/stroke risk prediction in DFI patients, which combines covariates such as office and laboratory-based biomarkers, carotid ultrasound image phenotype (CUSIP) lesions, along with the DFI severity. We confirmed the viability of CVD/stroke risk stratification in the DFI patients. Strong designs were found in the research of the DL architectures for CVD/stroke risk stratification. Finally, we analyzed the AI bias and proposed strategies for the early diagnosis of CVD/stroke in DFI patients. Since DFI patients have an aggressive atherosclerotic disease, leading to prominent CVD/stroke risk, we, therefore, conclude that the DL paradigm is very effective for predicting the risk of CVD/stroke in DFI patients.

Diagnosis and treatment in chronic pancreatitis: an international survey and case vignette study

BACKGROUND

The aim of the study was to evaluate the current opinion and clinical decision-making process of international pancreatologists, and to systematically identify key study questions regarding the diagnosis and treatment of chronic pancreatitis (CP) for future research.

METHODS

An online survey, including questions regarding the diagnosis and treatment of CP and several controversial clinical case vignettes, was send by e-mail to members of various international pancreatic associations: IHPBA, APA, EPC, ESGE and DPSG.

RESULTS

A total of 288 pancreatologists, 56% surgeons and 44% gastroenterologists, from at least 47 countries, participated in the survey. About half (48%) of the specialists used a classification tool for the diagnosis of CP, including the Mayo Clinic (28%), Mannheim (25%), or Büchler (25%) tools. Overall, CT was the preferred imaging modality for evaluation of an enlarged pancreatic head (59%), pseudocyst (55%), calcifications (75%), and peripancreatic fat infiltration (68%). MRI was preferred for assessment of main pancreatic duct (MPD) abnormalities (60%). Total pancreatectomy with auto-islet transplantation was the preferred treatment in patients with parenchymal calcifications without MPD abnormalities and in patients with refractory pain despite maximal medical, endoscopic, and surgical treatment. In patients with an enlarged pancreatic head, 58% preferred initial surgery (PPPD) versus 42% initial endoscopy. In patients with a dilated MPD and intraductal stones 56% preferred initial endoscopic ± ESWL treatment and 29% preferred initial surgical treatment.

CONCLUSION

Worldwide, clinical decision-making in CP is largely based on local expertise, beliefs and disbeliefs. Further development of evidence-based guidelines based on well designed (randomized) studies is strongly encouraged.

Long-term outcome of a cuffed expanded PTFE graft for hemodialysis vascular access

BACKGROUND

Development of venous outflow stenosis has been a major obstacle in maintaining the patency of dialysis vascular grafts. In the present study, we retrospectively analyzed the long-term results of cuffed expanded polytetrafluoroethylene (ePTFE) and non-cuffed (standard) ePTFE grafts placed for hemodialysis access.

MATERIALS AND METHODS

A total of 67 patients who underwent placement of either cuffed (n = 41) or standard (n = 26) ePTFE grafts were retrospectively analyzed. There were no significant differences between the two groups with regard to age, gender, cause of end-stage renal disease, and anatomic placement of the graft. Endpoints consisted of primary (without any intervention) and secondary (with radiological or surgical intervention) graft patency rates at 1-3 years.

RESULTS

There was a trend toward better primary graft patency rates in the cuffed versus the standard ePTFE: 37.7% vs. 25.7% at 1 year, 35% vs. 10.3% at 2 years, 28% vs. 5.1% at 3 years, respectively (p = 0.086, Kaplan-Meier). Secondary patency rates in the cuffed group were significantly superior (p = 0.047) to those in the standard group (81.8% vs. 56.1% at 1 year, 61.8% vs. 46.3% at 2 years, 51.5% vs. 33.1% at 3 years, respectively). Thrombosis as a cause of complete graft failure was significantly higher (34%) in the standard group than in the cuffed group (9%) (p = 0.0125).

CONCLUSIONS

Compared to the standard ePTFE, the cuffed ePTFE graft provided better long-term outcome, especially in terms of secondary patency rates after radiological intervention.

Frailty assessment and postoperative outcomes among patients undergoing general surgery

BACKGROUND

Frailty is an emerging concept in modern general surgery because of its correlation with adverse outcomes. More frail older patients are undergoing general surgery due to the rapid aging of the population and the effect of the "baby boom" generation. However, there is no consensus on the definition of frailty and on ways to assess its severity and effect.

PURPOSE

To describe the definition and epidemiology, measurement tools, and the effect of frailty on postoperative outcomes after general surgery.

METHODS

PubMed and Google Scholar databases were comprehensively searched.

RESULTS

Frailty is a syndrome defined as increased vulnerability to stressors due to a decline in physiological function and reserve among organ systems, resulting in adverse outcomes. Numerous tools have been described and tested for frailty measurement, but the ideal clinical tool has not been found yet. The evidence from cohort studies and meta-analyses shows associations between preoperative frailty and adverse perioperative outcomes after general surgery.

CONCLUSION

Frailty is an essential concept in general surgery. However, further studies have to identify the optimal way to preoperatively assess frailty and risk-stratify older patients.

Death following pulmonary complications of surgery before and during the SARS-CoV-2 pandemic

BACKGROUND

This study aimed to determine the impact of pulmonary complications on death after surgery both before and during the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) pandemic.

METHODS

This was a patient-level, comparative analysis of two, international prospective cohort studies: one before the pandemic (January-October 2019) and the second during the SARS-CoV-2 pandemic (local emergence of COVID-19 up to 19 April 2020). Both included patients undergoing elective resection of an intra-abdominal cancer with curative intent across five surgical oncology disciplines. Patient selection and rates of 30-day postoperative pulmonary complications were compared. The primary outcome was 30-day postoperative mortality. Mediation analysis using a natural-effects model was used to estimate the proportion of deaths during the pandemic attributable to SARS-CoV-2 infection.

RESULTS

This study included 7402 patients from 50 countries; 3031 (40.9 per cent) underwent surgery before and 4371 (59.1 per cent) during the pandemic. Overall, 4.3 per cent (187 of 4371) developed postoperative SARS-CoV-2 in the pandemic cohort. The pulmonary complication rate was similar (7.1 per cent (216 of 3031) versus 6.3 per cent (274 of 4371); P = 0.158) but the mortality rate was significantly higher (0.7 per cent (20 of 3031) versus 2.0 per cent (87 of 4371); P < 0.001) among patients who had surgery during the pandemic. The adjusted odds of death were higher during than before the pandemic (odds ratio (OR) 2.72, 95 per cent c.i. 1.58 to 4.67; P < 0.001). In mediation analysis, 54.8 per cent of excess postoperative deaths during the pandemic were estimated to be attributable to SARS-CoV-2 (OR 1.73, 1.40 to 2.13; P < 0.001).

CONCLUSION

Although providers may have selected patients with a lower risk profile for surgery during the pandemic, this did not mitigate the likelihood of death through SARS-CoV-2 infection. Care providers must act urgently to protect surgical patients from SARS-CoV-2 infection.

Promoting Undergraduate Surgical Education: Current Evidence and Students' Views on ESMSC International Wet Lab Course

BACKGROUND

Undergraduate Surgical Education is becoming an essential element in the training of the future generation of safe and efficient surgeons. Essential Skills in the Management of Surgical Cases (ESMSC), is an international, joint applied surgical science and simulation-based learning wet lab course.

METHODS

We performed a review of the existing literature on the topic of undergraduate surgical education. Following that, we analyzed the feedback questionnaire received 480 from 2 recent series of ESMSC courses (May 2015, n = 49 and November 2015, n = 40), in order to evaluate European Union students' (UK, Germany, Greece) views on the ESMSC course, as well as on the undergraduate surgical education. Results Using a 10 point graded scale, the overall ESMSC concept was positively evaluated, with a mean score of 9.41 ± 0.72 (range: 8-10) and 8.94 ± 1.1 (range: 7-10). The majority of delegates from both series [9.86 ± 0.43 (range: 8-10) and 9.58 ± 0.91 (range: 6-10), respectively] believed that ESMSC should be incorporated in the undergraduate surgical curriculum. Comparison of responses from the UK to the Greek Medical Student, as well as the findings from the third and fourth year versus the fifth and sixth year Medical Students, revealed no statistically significant differences pertaining to any of the questions (p > 0.05).

CONCLUSIONS

Current evidence in the literature supports the enhancement of surgical education through the systematic use of various modalities that provide Simulation-Based Training (SBT) hands-on experience, starting from the early undergraduate level. The findings of the present study are in agreement with these previous reports.

Vascular Implications of COVID-19: Role of Radiological Imaging, Artificial Intelligence, and Tissue Characterization: A Special Report

The SARS-CoV-2 virus has caused a pandemic, infecting nearly 80 million people worldwide, with mortality exceeding six million. The average survival span is just 14 days from the time the symptoms become aggressive. The present study delineates the deep-driven vascular damage in the pulmonary, renal, coronary, and carotid vessels due to SARS-CoV-2. This special report addresses an important gap in the literature in understanding (i) the pathophysiology of vascular damage and the role of medical imaging in the visualization of the damage caused by SARS-CoV-2, and (ii) further understanding the severity of COVID-19 using artificial intelligence (AI)-based tissue characterization (TC). PRISMA was used to select 296 studies for AI-based TC. Radiological imaging techniques such as magnetic resonance imaging (MRI), computed tomography (CT), and ultrasound were selected for imaging of the vasculature infected by COVID-19. Four kinds of hypotheses are presented for showing the vascular damage in radiological images due to COVID-19. Three kinds of AI models, namely, machine learning, deep learning, and transfer learning, are used for TC. Further, the study presents recommendations for improving AI-based architectures for vascular studies. We conclude that the process of vascular damage due to COVID-19 has similarities across vessel types, even though it results in multi-organ dysfunction. Although the mortality rate is ~2% of those infected, the long-term effect of COVID-19 needs monitoring to avoid deaths. AI seems to be penetrating the health care industry at warp speed, and we expect to see an emerging role in patient care, reduce the mortality and morbidity rate.

Long-term experience with liver transplantation for hepatocellular carcinoma

BACKGROUND

The effect of the model for end-stage liver disease (MELD) system on the post-transplant survival of patients with hepatocellular carcinoma (HCC) has not been fully elucidated. Our objective is to review the results of liver transplantation (LT) for HCC at the Massachusetts General Hospital over a period of 12 years, with special emphasis on the effect of the MELD system.

METHODS

A retrospective review of 73 patients who underwent liver transplantation for HCC between 1995 and 2007. Outcome measures included demographics, tumor stage at explant, patient survival, and tumor recurrence.

RESULTS

On pathologic review of the explanted liver, 12.3% of patients were classified as stage I; 42.5% as stage II, 21.9% as stage III, and 23.3% as stage IV. The overall actual survival rate was 85% at 1 year, 82% at 3 years, 73% at 5 years, and 66% at 10 years. Overall tumor recurrence was 11%. Survival rates were higher after the MELD system (5 year survival 60% before MELD vs. 85% after MELD). Recurrence decreased from 21% to 7.5%.

CONCLUSION

We showed improved survival for HCC after LT over the last 12 years, and especially improved survival and decreased recurrence in the time since the implementation of the MELD system.

Hydrodynamic plasmid DNA gene therapy model in liver transplantation

BACKGROUND

There is great interest in the field of transplantation to genetically modify grafts to decrease preservation injury or allograft rejection. Although adenoviral gene transfer has been effective in experimental liver transplantation, viral toxicity and safety concerns limit potential use in clinical trials. Therefore, the purpose of this study was to develop a model of nonviral gene transfer in the liver transplant setting, allowing for efficient transgene expression.

MATERIALS AND METHODS

Orthotopic syngeneic rat liver transplantation was performed with 3 h cold ischemia using University of Wisconsin (UW) preservation. A hydrodynamic gene transfer technique was developed where plasmid DNA was delivered to the liver graft by ex vivo rapid infusion of DNA in UW via the IVC with other vessels clamped. Expression plasmids for the marker genes luciferase and secreted human alpha1-antitrypsin (alpha1-AT) were used. Hepatic injury was assessed by graft histology and liver transaminases. Transgene expression was determined by hepatic luciferase relative light units activity (RLU) and serum alpha1-AT protein levels. Variables examined included the effect of (a) volume injected on the intravenous pressure in the liver graft; (b) injury to the liver, as measured by hepatic enzymes and histopathology; (c) variable expression between lobes; (d) volume of UW that the plasmid is administered in; (e) amount of DNA plasmid; (f) type of the promoter used; (g) clamp time; as well as (h) the time course of the marker gene expression.

RESULTS

Control rats underwent standard orthotopic syngeneic rat liver transplantation and had no detectable hepatic luciferase activity or serum human alpha1-AT. The optimal DNA plasmid dose was found to be 400 mug/liver graft, as there was no increase in the luciferase expression by increasing the dose. Furthermore, cytomegalovirus promoter yielded greater expression than Rous sarcoma virus. A high injection pressure gradient allowed for more efficient transgene expression, but produced greater liver injury shown by elevated transaminases and centrilobular necrosis. Lowering injection volume from 75 to 50% of graft weight decreased liver injury by 4.5-fold. Although higher UW injection volumes were associated with increased expression, volumes of only 50% led to luciferase expression up to 10,000,000 RLU/mg; this expression was homogeneous between the different liver lobes. Human alpha1-AT was detected in recipient blood as early as 6 h, peaked at 24 h, and remained high for 5 days.

CONCLUSIONS

We have developed a nonviral gene transfer technique where hydrostatic pressure across the cold-preserved liver vascular bed allows for efficient plasmid DNA delivery. This simple strategy should prove useful to genetically modify liver grafts in the transplantation setting.

COVID-19-related absence among surgeons: development of an international surgical workforce prediction model

BACKGROUND

During the initial COVID-19 outbreak up to 28.4 million elective operations were cancelled worldwide, in part owing to concerns that it would be unsustainable to maintain elective surgery capacity because of COVID-19-related surgeon absence. Although many hospitals are now recovering, surgical teams need strategies to prepare for future outbreaks. This study aimed to develop a framework to predict elective surgery capacity during future COVID-19 outbreaks.

METHODS

An international cross-sectional study determined real-world COVID-19-related absence rates among surgeons. COVID-19-related absences included sickness, self-isolation, shielding, and caring for family. To estimate elective surgical capacity during future outbreaks, an expert elicitation study was undertaken with senior surgeons to determine the minimum surgical staff required to provide surgical services while maintaining a range of elective surgery volumes (0, 25, 50 or 75 per cent).

RESULTS

Based on data from 364 hospitals across 65 countries, the COVID-19-related absence rate during the initial 6 weeks of the outbreak ranged from 20.5 to 24.7 per cent (mean average fortnightly). In weeks 7-12, this decreased to 9.2-13.8 per cent. At all times during the COVID-19 outbreak there was predicted to be sufficient surgical staff available to maintain at least 75 per cent of regular elective surgical volume. Overall, there was predicted capacity for surgeon redeployment to support the wider hospital response to COVID-19.

CONCLUSION

This framework will inform elective surgical service planning during future COVID-19 outbreaks. In most settings, surgeon absence is unlikely to be the factor limiting elective surgery capacity.

Laparoscopic versus open hepatectomy for intrahepatic cholangiocarcinoma: An individual patient data survival meta-analysis

BACKGROUND

We compared the outcomes of laparoscopic hepatectomy (LH) vs. open hepatectomy (OH) for intrahepatic cholangiocarcinoma (iCCA).

METHODS

A systematic review of the MEDLINE, EMBASE, Scopus, and Cochrane Library databases was performed using PRISMA guidelines (end-of-search date: 08-June-2020). Individual patient data on overall survival (OS) and recurrence-free survival (RFS) were extracted. Random-effects meta-analyses, and one- and two-stage survival analyses were conducted.

RESULTS

Eight retrospective cohort studies comparing LH (n = 544) vs. OH (n = 2256) were identified. LH demonstrated lower overall complication (Risk ratio [RR] = 0.64, 95% confidence interval [CI]: 0.46-0.90; p = 0.01), surgical lymphadenectomy (RR = 0.74, 95% CI: 0.58-0.93; p = 0.01) and margin-positive resection (RR = 0.78, 95% CI: 0.62-0.99; p = 0.04) rates, and higher recurrence-free rate (RR = 1.24, 95% CI: 1.01-1.51; p = 0.04) vs. OH. In Cox regression, no difference was observed regarding OS (Hazard Ratio [HR] = 1.11, 95% CI: 0.65-1.91; p = 0.70) and RFS (HR = 1.19, 95% CI: 0.74-1.90; p = 0.47).

CONCLUSION

The use of LH should be considered when feasible in well-selected iCCA patients by hepatobiliary surgeons with experience in minimally-invasive surgery.

Pharmacokinetics of tacrolimus in living donor liver transplant and deceased donor liver transplant recipients

INTRODUCTION

Hepatic dysfunction is an important determinant of the clearance of tacrolimus; however, the impact of reduced hepatic mass in living donor liver transplant (LDLT) patients on the drug exposure and clearance of tacrolimus is not known. AIM.: The aim of the present study is to compare the dosage, concentration and pharmacokinetics parameters of tacrolimus between LDLT and deceased donor liver transplant (DDLT) recipients.

PATIENTS AND METHODS

Daily doses used and trough concentrations measured were compared in 12 LDLT and 12 DDLT patients. Multiple blood samples were taken over one dosing interval after oral tacrolimus administration, and pharmacokinetics differences were compared.

RESULTS

The mean tacrolimus dosage in first 14 postoperative days was (0.06 mg/kg/day) for LDLT and (0.09 mg/kg/day) for DDLT (P=0.0001). Despite the lower doses used, mean trough concentration was significantly greater in LDLT as compared with DDLT (8.8+/-2.5 ng/mL vs. 6.79+/-1.5 ng/mL, respectively, P=0.013). On the day of the pharmacokinetic study, minimum Concentration (Cmin), 12-hr postdose concentration (Clast), and average concentration (Cavg) were significantly greater in LDLT as compared with DDLT (LDLT: 6.6+/-2.4 ng/mL, 7.2+/-1.8 ng/mL, 8.9+/-3.0 ng/mL; DDLT: 4.3+/-1.0 ng/mL, 4.9+/-1.6 ng/mL, 5.9+/-1.4 ng/mL, P=0.02, 0.04, and 0.02, respectively). Dose normalized AUC was 37.7% greater and clearance, 47.5% lower in LDLT as compared with DDLT.

CONCLUSION

Although not statistically significant, the dose normalized AUC was 37.7% greater and clearance 47.5% lower in LDLT as compared with DDLT. An initial tacrolimus dose reduction of about 30-40% may be prudent in LDLT compared with DDLT recipients.

Developing a novel international undergraduate surgical masterclass during a financial crisis: our 4-year experience

BACKGROUND

Essential Skills in the Management of Surgical Cases (ESMSC) is a novel 3-day international undergraduate surgical masterclass. Its current curriculum (Cores integrated for Research-Ci4R) is built on a tetracore, multiclustered architecture combining high-fidelity and low-fidelity simulation-based learning (SBL), with applied and basic science case-based workshops, and non-technical skills modules. We aimed to report our experience in setting up ESMSC during the global financial crisis.

METHODS

We report the evolution of our curriculum's methodology and summarised the research outcomes related to the objective performance improvement of delegates, the educational environment of the course and the use of mixed-fidelity SBL. Feedback from the last three series of the course was prospectively collected and analysed using univariate statistics on IBM SPSS V.23.

RESULTS

311 medical students across the European Union (EU) were selected from a competitive pool of 1280 applicants during seven series of the course between 2014 and 2017. During this period, curriculum 14 s evolved to the final Ci4R version, which integrates a tetracore structure combining 32 stations of in vivo, ex vivo and dry lab SBL with small group teaching workshops. Ci4R was positively perceived across different educational background students (p>0.05 for any comparison).

CONCLUSIONS

ESMSC is considered an innovative and effective multidisciplinary teaching model by delegates, where it improves delegates objective performance in basic surgical skills. Our experience demonstrates provision of high-quality and free surgical education during a financial crisis, which evolved through a dynamic feedback mechanism. The prospective recording and subsequent analysis of curriculum evolution provides a blueprint to direct development of effective surgical education courses that can be adapted to local needs.

Prognostic value of COL6A3 in pancreatic adenocarcinoma

BACKGROUNDS/AIMS

Pancreatic cancer is one of the most fatal human malignancies with poor prognosis, despite advances in therapy. Here, we evaluated the potential role of collagen type VI α3 chain (COL6A3) as a non-invasive biomarker for pancreatic adenocarcinoma.

METHODS

In this study, we investigated immunohistochemically the expression of COL6A3 in 30 patients with resectable pancreatic adenocarcinoma by immunohistochemistry in a tissue sample of the cancer and a tissue sample of normal pancreas for each patient. Also, we looked for associations between COL6A3 and other prognostic factors of pancreatic cancer.

RESULTS

All of the pancreatic cancer tissue samples revealed in different ranges of intensity from weak (+) in 16.67%, moderate (+2) in 50%, to strongly positive (+3) in 33.33% staining for COL6A3. We found no moderate or strongly positive staining in normal pancreatic tissue. There was only weak positive staining in 23 samples (76.67%) and 7 (23.30%) were negative. Also, there was significant correlation between COL6A3 moderate and strongly expression and negative prognostic factors for pancreatic cancer.

CONCLUSIONS

The greatest density of COL6A3 was observed in pancreatic cancer tissues and was correlated with negative prognostic factors for pancreatic cancer. Therefore, we suggest that COL6A3 could be used as prognostic factor in pancreatic cancer, but more studies need to prove its value.

COVLIAS 1.0Lesion vs. MedSeg: An Artificial Intelligence Framework for Automated Lesion Segmentation in COVID-19 Lung Computed Tomography Scans

Background: COVID-19 is a disease with multiple variants, and is quickly spreading throughout the world. It is crucial to identify patients who are suspected of having COVID-19 early, because the vaccine is not readily available in certain parts of the world. Methodology: Lung computed tomography (CT) imaging can be used to diagnose COVID-19 as an alternative to the RT-PCR test in some cases. The occurrence of ground-glass opacities in the lung region is a characteristic of COVID-19 in chest CT scans, and these are daunting to locate and segment manually. The proposed study consists of a combination of solo deep learning (DL) and hybrid DL (HDL) models to tackle the lesion location and segmentation more quickly. One DL and four HDL models—namely, PSPNet, VGG-SegNet, ResNet-SegNet, VGG-UNet, and ResNet-UNet—were trained by an expert radiologist. The training scheme adopted a fivefold cross-validation strategy on a cohort of 3000 images selected from a set of 40 COVID-19-positive individuals. Results: The proposed variability study uses tracings from two trained radiologists as part of the validation. Five artificial intelligence (AI) models were benchmarked against MedSeg. The best AI model, ResNet-UNet, was superior to MedSeg by 9% and 15% for Dice and Jaccard, respectively, when compared against MD 1, and by 4% and 8%, respectively, when compared against MD 2. Statistical tests—namely, the Mann−Whitney test, paired t-test, and Wilcoxon test—demonstrated its stability and reliability, with p < 0.0001. The online system for each slice was <1 s. Conclusions: The AI models reliably located and segmented COVID-19 lesions in CT scans. The COVLIAS 1.0Lesion lesion locator passed the intervariability test.

Inter-Variability Study of COVLIAS 1.0: Hybrid Deep Learning Models for COVID-19 Lung Segmentation in Computed Tomography

Background: For COVID-19 lung severity, segmentation of lungs on computed tomography (CT) is the first crucial step. Current deep learning (DL)-based Artificial Intelligence (AI) models have a bias in the training stage of segmentation because only one set of ground truth (GT) annotations are evaluated. We propose a robust and stable inter-variability analysis of CT lung segmentation in COVID-19 to avoid the effect of bias. Methodology: The proposed inter-variability study consists of two GT tracers for lung segmentation on chest CT. Three AI models, PSP Net, VGG-SegNet, and ResNet-SegNet, were trained using GT annotations. We hypothesized that if AI models are trained on the GT tracings from multiple experience levels, and if the AI performance on the test data between these AI models is within the 5% range, one can consider such an AI model robust and unbiased. The K5 protocol (training to testing: 80%:20%) was adapted. Ten kinds of metrics were used for performance evaluation. Results: The database consisted of 5000 CT chest images from 72 COVID-19-infected patients. By computing the coefficient of correlations (CC) between the output of the two AI models trained corresponding to the two GT tracers, computing their differences in their CC, and repeating the process for all three AI-models, we show the differences as 0%, 0.51%, and 2.04% (all < 5%), thereby validating the hypothesis. The performance was comparable; however, it had the following order: ResNet-SegNet > PSP Net > VGG-SegNet. Conclusions: The AI models were clinically robust and stable during the inter-variability analysis on the CT lung segmentation on COVID-19 patients.

Successful, combined long-term treatment of cerebral aspergillosis in a liver transplant patient

Invasive aspergillosis has long been recognized as one of the most significant and often fatal opportunistic fungal infections in liver transplant recipients. We report a case of a liver transplant recipient who developed an Aspergillus fumigatus brain abscess that produced significant neurologic symptoms. The patient was managed successfully with a combination of surgery and medical treatment with Voriconazole. To our knowledge, this is the second such case reported in the literature.

The process of estimating the cost of surgery: Providing a practical framework for surgeons

Over the last decades, health care costs have been increasing at an alarming, exponential rate which is considered unsustainable. Surgical care utilizes one-third of health care costs. Estimating, evaluating, and understanding the cost of surgery is a vital step towards cost management and reduction. Current cost estimation studies and cost-effectiveness studies have vast disparities in their methodology, with published costs of Operating Room varying from as low as $7 and as high as $113 per minute. Costs in surgery are distinguished as direct and indirect. Allocation of direct costs involves identification, measurement, and valuation processes. Allocation of indirect costs involves the allocation of capital and overhead costs and of indirect department costs. Annualised capital costs and overhead hospital costs are then allocated to surgery by either the cost-centre allocation or the activity-based allocation frameworks. Indirect department costs are allocated to a specific surgery by weighted service allocation or hourly rate allocation or inpatient day allocation, or marginal markup allocation. The growing societal, financial and political pressure for cost reduction has brought cost analysis to the forefront of healthcare discussions. Thus, we believe that almost every single surgeon will eventually enter the field of healthcare economics by necessity. This review aims to provide surgeons with a practical framework for engaging in cost estimation studies.

Nutrition, atherosclerosis, arterial imaging, cardiovascular risk stratification, and manifestations in COVID-19 framework: a narrative review

Background: Atherosclerosis is the primary cause of the cardiovascular disease (CVD). Several risk factors lead to atherosclerosis, and altered nutrition is one among those. Nutrition has been ignored quite often in the process of CVD risk assessment. Altered nutrition along with carotid ultrasound imaging-driven atherosclerotic plaque features can help in understanding and banishing the problems associated with the late diagnosis of CVD. Artificial intelligence (AI) is another promisingly adopted technology for CVD risk assessment and management. Therefore, we hypothesize that the risk of atherosclerotic CVD can be accurately monitored using carotid ultrasound imaging, predicted using AI-based algorithms, and reduced with the help of proper nutrition. Layout: The review presents a pathophysiological link between nutrition and atherosclerosis by gaining a deep insight into the processes involved at each stage of plaque development. After targeting the causes and finding out results by low-cost, user-friendly, ultrasound-based arterial imaging, it is important to (i) stratify the risks and (ii) monitor them by measuring plaque burden and computing risk score as part of the preventive framework. Artificial intelligence (AI)-based strategies are used to provide efficient CVD risk assessments. Finally, the review presents the role of AI for CVD risk assessment during COVID-19. Conclusions: By studying the mechanism of low-density lipoprotein formation, saturated and trans fat, and other dietary components that lead to plaque formation, we demonstrate the use of CVD risk assessment due to nutrition and atherosclerosis disease formation during normal and COVID times. Further, nutrition if included, as a part of the associated risk factors can benefit from atherosclerotic disease progression and its management using AI-based CVD risk assessment.

Potential immunological advantage of intravenous mycophenolate mofetil with tacrolimus and steroids in primary deceased donor liver transplantation and live donor liver transplantation without antibody induction

With the current immunosuppressive regimens, graft loss secondary to immunological reasons after successful liver transplantation is a rarity; acute rejections, however, do occur, with the majority of them being steroid-responsive. The aim of the present study is to examine the rate of acute rejection with tacrolimus, intravenous (IV) mycophenolate mofetil (MMF), and steroids in primary deceased donor liver transplant (DDLT) and live donor liver transplant (LDLT) recipients. During the year 2005, 130 patients (mean age: 54.9 +/- 10.8, males: 84, females: 46, 112 DDLT and 18 LDLT) received primary liver transplantation. They were followed up for the incidence of acute rejection in the first 12 months. Liver biopsies were performed as clinically indicated; protocol liver biopsies were never performed. A total of 127 liver biopsies were performed. Thirty-two had a rejection activity index (RAI) score of > or =3, of which 24 biopsies in 20 patients were not treated with a steroid bolus. Eight (6.1%) patients (mean RAI score: 5.1 +/- 1.4) received 750 to 1500 mg of methylprednisolone over 3 days. Out of these, 2 were noncompliant, 4 were off MMF, and 1 was on cyclosporine. All patients responded to steroid therapy. None of the patients required any antibody preparation. In conclusion, IV MMF with tacrolimus and steroids is useful and required antirejection therapy in 6.1% of liver transplant recipients.

COVIDSurg Collaborative. The impact of surgical delay on resectability of colorectal cancer: An international prospective cohort study

AIM

The SARS-CoV-2 pandemic has provided a unique opportunity to explore the impact of surgical delays on cancer resectability. This study aimed to compare resectability for colorectal cancer patients undergoing delayed versus non-delayed surgery.

METHODS

This was an international prospective cohort study of consecutive colorectal cancer patients with a decision for curative surgery (January-April 2020). Surgical delay was defined as an operation taking place more than 4 weeks after treatment decision, in a patient who did not receive neoadjuvant therapy. A subgroup analysis explored the effects of delay in elective patients only. The impact of longer delays was explored in a sensitivity analysis. The primary outcome was complete resection, defined as curative resection with an R0 margin.

RESULTS

Overall, 5453 patients from 304 hospitals in 47 countries were included, of whom 6.6% (358/5453) did not receive their planned operation. Of the 4304 operated patients without neoadjuvant therapy, 40.5% (1744/4304) were delayed beyond 4 weeks. Delayed patients were more likely to be older, men, more comorbid, have higher body mass index and have rectal cancer and early stage disease. Delayed patients had higher unadjusted rates of complete resection (93.7% vs. 91.9%, P = 0.032) and lower rates of emergency surgery (4.5% vs. 22.5%, P < 0.001). After adjustment, delay was not associated with a lower rate of complete resection (OR 1.18, 95% CI 0.90-1.55, P = 0.224), which was consistent in elective patients only (OR 0.94, 95% CI 0.69-1.27, P = 0.672). Longer delays were not associated with poorer outcomes.

CONCLUSION

One in 15 colorectal cancer patients did not receive their planned operation during the first wave of COVID-19. Surgical delay did not appear to compromise resectability, raising the hypothesis that any reduction in long-term survival attributable to delays is likely to be due to micro-metastatic disease.

Systematic review and meta-analysis of open versus laparoscopy-assisted versus pure laparoscopic versus robotic living donor hepatectomy

The value of minimally invasive approaches for living donor hepatectomy remains unclear. Our aim was to compare the donor outcomes after open versus laparoscopy-assisted versus pure laparoscopic versus robotic living donor hepatectomy (OLDH vs. LALDH vs. PLLDH vs. RLDH). A systematic literature review of the MEDLINE, Cochrane Library, Embase, and Scopus databases was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) statement (up to December 8, 2021). Random-effects meta-analyses were performed separately for minor and major living donor hepatectomy. The risk of bias in nonrandomized studies was assessed using the Newcastle-Ottawa Scale. A total of 31 studies were included. There was no difference in donor outcomes after OLDH versus LALDH for major hepatectomy. However, PLLDH was associated with decreased estimated blood loss, length of stay (LOS), and overall complications versus OLDH for minor and major hepatectomy, but also with increased operative time for major hepatectomy. PLLDH was associated with decreased LOS versus LALDH for major hepatectomy. RLDH was associated with decreased LOS but with increased operative time versus OLDH for major hepatectomy. The scarcity of studies comparing RLDH versus LALDH/PLLDH did not allow us to meta-analyze donor outcomes for that comparison. There seems to be a marginal benefit in estimated blood loss and/or LOS in favor of PLLDH and RLDH. The complexity of these procedures limits them to transplant centers with high volume and experience. Future studies should investigate self-reported donor experience and the associated economic costs of these approaches.

Is In-Vivo laparoscopic simulation learning a step forward in the Undergraduate Surgical Education?

BACKGROUND

Essentials Skills in the Management of Surgical Cases - ESMSC is an International Combined Applied Surgical Science and Wet Lab course addressed at the Undergraduate level. Laparoscopic Skills is a fundamental element of Surgical Education and various Simulation-Based Learning (SBL) models have been endorsed. This study aims to explore if there is any significant difference in delegates' performance depending on whether they completed In Vivo module prior to the equivalent in the laparoscopic simulator.

MATERIALS AND METHODS

37 Medical Students from various EU countries were divided in 2 groups, and both completed the "Fundamentals in Laparoscopic Surgery" module in the Dry-lab Laparoscopic Simulator as well as the same module "In Vivo" on a swine model. Group A (18 students, 48.6%) completed the "Fundamentals in Laparoscopic Surgery - FLS" module prior to the "In Vivo", whereas group B completed the "In Vivo" module first. Direct Observation of Procedural Skills (DOPS) were used to assess delegates' performance.

RESULTS

The mean DOPS scores for the "FLS" and "In Vivo" models were 2.27 ± 0.902 and 2.03 ± 0.833, respectively, and the delegates' performance was not statistically significantly different between them (p = 0.128). There was no statistically significant difference in the scores among different gender, year of study, school and handedness groups. The alteration in the sequence between Dry-lab "FLS" and "In Vivo" modules did not affect the performance in neither the "FLS" nor the "In Vivo" models.

CONCLUSIONS

The inexpensive, but low-fidelity "FLS" model could serve an equal alternative Simulation-Based Learning model for the early undergraduate training. Our study demonstrated that high fidelity In Vivo simulation for laparoscopic skills does not affect significantly the improvement in the delegates' performance at the undergraduate level. Further studies should be conducted to identify at which stage of training should high fidelity simulation be introduced.

Ensemble Deep Learning Derived from Transfer Learning for Classification of COVID-19 Patients on Hybrid Deep-Learning-Based Lung Segmentation: A Data Augmentation and Balancing Framework

BACKGROUND AND MOTIVATION

Lung computed tomography (CT) techniques are high-resolution and are well adopted in the intensive care unit (ICU) for COVID-19 disease control classification. Most artificial intelligence (AI) systems do not undergo generalization and are typically overfitted. Such trained AI systems are not practical for clinical settings and therefore do not give accurate results when executed on unseen data sets. We hypothesize that ensemble deep learning (EDL) is superior to deep transfer learning (TL) in both non-augmented and augmented frameworks.

METHODOLOGY

The system consists of a cascade of quality control, ResNet-UNet-based hybrid deep learning for lung segmentation, and seven models using TL-based classification followed by five types of EDL's. To prove our hypothesis, five different kinds of data combinations (DC) were designed using a combination of two multicenter cohorts-Croatia (80 COVID) and Italy (72 COVID and 30 controls)-leading to 12,000 CT slices. As part of generalization, the system was tested on unseen data and statistically tested for reliability/stability.

RESULTS

Using the K5 (80:20) cross-validation protocol on the balanced and augmented dataset, the five DC datasets improved TL mean accuracy by 3.32%, 6.56%, 12.96%, 47.1%, and 2.78%, respectively. The five EDL systems showed improvements in accuracy of 2.12%, 5.78%, 6.72%, 32.05%, and 2.40%, thus validating our hypothesis. All statistical tests proved positive for reliability and stability.

CONCLUSION

EDL showed superior performance to TL systems for both (a) unbalanced and unaugmented and (b) balanced and augmented datasets for both (i) seen and (ii) unseen paradigms, validating both our hypotheses.

Retrospective qualitative study evaluating the application of IG4 curriculum: an adaptable concept for holistic surgical education

OBJECTIVES

Faced with a costly and demanding learning curve of surgical skills acquisition, the growing necessity for improved surgical curricula has now become irrefutable. We took this opportunity to formulate a teaching framework with the capacity to provide holistic surgical education at the undergraduate level.

SETTING

Data collection was conducted in all the relevant healthcare centres the participants worked in. Where this was not possible, interviews were held in quiet public places.

PARTICIPANTS

We performed an in-depth retrospective evaluation of a proposed curriculum, through semi-structured interviews with 10 participants. A targeted sampling technique was employed in order to identify senior academics with specialist knowledge in surgical education. Recruitment was ceased on reaching data saturation after which thematic data analysis was performed using NVivo 11.

RESULTS

Thematic analysis yielded a total of 4 main themes and 29 daughter nodes. Majority of study participants agreed that the current landscape of basic surgical education is deficient at multiple levels. While simulation cannot replace surgical skills acquisition taking place in operating rooms, it can be catalytic in the transition of students to postgraduate training. Our study concluded that a standardised format of surgical teaching is essential, and that the Integrated Generation 4 (IG4) framework provides an excellent starting point.

CONCLUSIONS

Through expert opinion, IG4 has been validated for its capacity to effectively accommodate learning in a safer and more efficacious environment. Moreover, we support that through dissemination of IG4, we can instil a sense of motivation to students as well as develop robust data sets, which will be amenable to data analysis through the application of more sophisticated methodologies.