Search

Feature selection is essential in medical area; however, its process becomes complicated with the presence of censoring which is the unique character of survival analysis. Most survival feature selection methods are based on Cox's proportional hazard model, though machine learning classifiers are preferred. They are less employed in survival analysis due to censoring which prevents them from directly being used to survival data. Among the few work that employed machine learning classifiers, partial logistic artificial neural network with auto-relevance determination is a well-known method that deals with censoring and perform feature selection for survival data. However, it depends on data replication to handle censoring which leads to unbalanced and biased prediction results especially in highly censored data. Other methods cannot deal with high censoring. Therefore, in this article, a new hybrid feature selection method is proposed which presents a solution to high level censoring. It combines support vector machine, neural network, and K-nearest neighbor classifiers using simple majority voting and a new weighted majority voting method based on survival metric to construct a multiple classifier system. The new hybrid feature selection process uses multiple classifier system as a wrapper method and merges it with iterated feature ranking filter method to further reduce features. Two endovascular aortic repair datasets containing 91% censored patients collected from two centers were used to construct a multicenter study to evaluate the performance of the proposed approach. The results showed the proposed technique outperformed individual classifiers and variable selection methods based on Cox's model such as Akaike and Bayesian information criterions and least absolute shrinkage and selector operator in p values of the log-rank test, sensitivity, and concordance index. This indicates that the proposed classifier is more powerful in correctly predicting the risk of re-intervention enabling doctor in selecting patients' future follow-up plan.

Approximately 40-50% of patients undergoing thoracic endovascular aortic repair require left subclavian artery coverage for adequate proximal landing zone. Many of these patients undergo left subclavian artery revascularisation. However, outcomes data for left subclavian artery revascularisation in the context of thoracic endovascular aortic repair remain limited. In this study, 70 left subclavian artery revascularisation procedures, performed on thoracic endovascular aortic repair patients at a tertiary hospital, were retrospectively reviewed. Particular emphasis was placed on revascularisation-related outcomes during staging interval between revascularisation and thoracic endovascular aortic repair. Forty-six (66%) carotid-subclavian bypass, 17 (24%) carotid-carotid-subclavian bypass and 7 (10%) aorto-inominate-carotid-subclavian bypass procedures were performed. There were no strokes or mortalities following left subclavian artery revascularisation procedures alone. Three (10%) minor complications occurred including a seroma, a haematoma and a temporary neuropraxia. Separation of complications following left subclavian artery revascularisation from those of the associated thoracic endovascular aortic repair can be difficult. Early outcomes data from patients who underwent left subclavian artery revascularisation in isolation indicate that the procedure is safe with low complication rates.

Objectives: Deep vein thrombosis (DVT) is a leading cause of cardiovascular disease. We tested the hypothesis that there is a consensus regarding the treatment of acute DVT among clinicians experienced in DVT management., Method: A Delphi consensus approach was used to gather expert opinion regarding attitudes towards the treatment of acute proximal DVT and management of specific cases. Strength of preference for various treatment strategies across a number of case scenarios was quantified. Univariate and multivariate analyses were performed to quantify the influence of various factors on treatment modality selected., Results: Respondents strongly agreed that DVT was a significant health problem and that further research was a priority. A multidisciplinary team approach with access to various treatment strategies was encouraged. Pregnancy and recent surgery independently predicted preference for medical treatment, whereas proximal DVT and May-Thurner syndrome were associated with interventional strategies., Conclusion: Acute proximal DVT is a significant health problem for which a variety of treatments are available. This study demonstrates that no consensus exists as to the optimum strategy for certain patient groups. Trends in opinion based on local experience and case-series exist, but the results of ongoing randomized trials will ultimately inform best practice.

This study aims to quantify the management and outcomes of patients treated for venous thoracic outlet syndrome (vTOS) over a seven-year period. A retrospective case-note review of all patients undergoing first rib resection at a regional vascular unit between January 1, 2002 and December 31, 2009 was performed. Treatment pathways and outcomes recorded as freedom from symptoms and venous patency were analyzed. Thirty-five patients were identified with vTOS. Ninety-one percent of patients had patent veins at discharge from clinical follow-up and were symptom-free at a median of 44 months. Patients treated within seven days of symptoms (94.7 versus 85.7, P = 0.060), with catheter-directed thrombolysis (94 versus 87.5% P = 0.702) and excision of first rib in less than 30 days (100 versus 85.7%, P = 0.002), had improved symptom-free rates. In conclusion, early referral, immediate catheter-directed thrombolysis, perioperative balloon venoplasty and early thoracic outlet decompression may provide a suitable model for improving outcomes in vTOS.

Objective: The aim of this study was to evaluate the training experience of current UK vascular trainees in the modern management of venous disease., Method: A web-based questionnaire of the 145 members of the Rouleaux Club (www.rouleauxclub.com), which represents UK vascular trainees. Members were asked to complete the survey between June and October 2009 with regular email reminders being sent out to non-responders., Results: One hundred and twenty-three trainees (85% response rate) representing all 17 UK training Deaneries responded. Seventy-eight per cent reported having received no formal venous duplex training either for diagnosis of venous disease or to guide endovenous therapy. Operative experience of great and small saphenous vein surgery improved with years of training. Surgical experience for recurrent varicose veins was poor. Experience with endovenous techniques was limited and variable. No experience of endovenous laser ablation or radiofrequency ablation was reported by 39% and 67% of trainees, respectively. Experience and/or training with foam sclerotherapy was limited to <40%. Many of those reporting no experience with endovenous ablation techniques were within the final two years of their training. Less than 25% of trainees reported having had any experience (assisted/performed) of advanced venous interventions such as thrombolysis techniques for deep venous thrombosis, inferior vena cava filter placement/removal, venous stenting or deep venous reconstruction. Less than a quarter of trainees are currently involved in the acute management of deep venous thrombosis. The majority (76%) of current trainees would like a formal approved UK venous training course to be offered., Conclusion: The current level of training in the management of venous disease will not allow UK vascular trainees to become the competent all round vascular specialists of the future.

Background: Deep vein thromboses (DVTs) are a significant cause of morbidity and mortality. Valvular destruction leads to the spectrum of disease called the post-thrombotic syndrome (PTS) with the sequelae of chronic venous ulceration and a reduced quality of life. Catheter-directed thrombolysis (CDT) may reduce the incidence of PTS following an acute proximal DVT and increases quality of life thereafter, but it is uncertain what proportion of patients diagnosed with a DVT would be suitable for CDT., Methods: This study quantified the proportion of patients investigated for DVT that would have been suitable for CDT. A retrospective review was performed of all upper and lower limb duplex ultrasound scans for suspected DVTs in a contemporary one-year period in a major regional vascular institute. All positive scans for acute proximal lower limb DVTs were compared against strict inclusion and exclusion criteria for CDT, based on national guidelines and international randomized trials., Results: A total of 2368 duplex ultrasound venous investigations were performed in a one-year period and 252 scans demonstrated DVT. Of these, 158 were acute proximal lower limb DVTs. Application of the inclusion and exclusion criteria for CDT suggested that 47/158 (30%) were potentially suitable for CDT using current criteria. The median age of the 158 patients was 58 years, meaning that more than half were of working age and 54% were men., Conclusion: Using current eligibility criteria, only about 30% of patients with DVT appear to be suitable for CDT.

The objectives of this study were to compare the 1-year patency of superficial femoral artery (SFA) stent placement with percutaneous transluminal angioplasty (PTA) alone and to attempt to define the role of stents in arterial occlusive disease. Literature searches of the Embase, Medline, and Cochrane databases identified relevant articles, which were split into two subgroups: those containing case-controlled matches for stenting and PTA and those considering only stent placement. The review conformed to the QUORUM statement. One-year patency rates were 219 of 383 (57%) in the stented group and 319 of 607 (53%) in the PTA group for matched cases (odds ratio 1.206 [95% CI 0.932-1.559; p = .115]). The patency of stents deployed secondarily was 554 of 909 (61%). The combined 1-year patency of primary and secondary SFA stents for matched and unmatched case series was 816 of 1,282 (64%). One-year patency rates following stent placement or PTA alone in well-matched patient groups demonstrated no significant difference. This would suggest that the routine use of primary SFA stenting should be undertaken only in selected cases and should mainly be used in "bailout" situations or for complex lesions where surgery is precluded. However, the studies used displayed a high degree of heterogeneity, and some used stent technology that is now considered obsolete. Ongoing randomized controlled trials will provide a more definitive answer to this important problem.

Endovascular techniques are well established in minimally invasive vascular surgery. Stent-graft technology is evolving. This will allow more complex aortic pathology to be treated. Current limitations in this equipment include the reliance on selective catheters along with trial and error to cannulate visceral or aortic arch branch vessels. This can lead to prolonged operations with increased radiation and contrast load to the patient with inevitable increase in peri-operative morbidity. A smart catheter that could expedite target vessel catheterization, minimize repeat catheter changes, and provide greater haptic feedback would enhance the endovascular surgeon's armamentarium. This would lead to broader and safer application of endovascular surgery.

Atherosclerotic plaques are a feature of abdominal aortic aneurysms (AAAs). Atherosclerosis and AAA appear to share similar risk factors. These observations have led to the conclusion that AAAs are a consequence of advanced atherosclerosis.This review explores current theories regarding the pathogenesis of AAA and their implications for treatment.A systematic literature search was conducted using the search terms abdominal aortic aneurysm, atherosclerosis, pathogenesis, and systemic disease. Articles were categorized according to the association of AAAs with atherosclerosis, arteriomegaly, peripheral aneurysm, systemic expression, genetics, autoimmunity, oxidative stress, and systemic disease. Twenty-nine articles reporting changes in the systemic vasculature associated with AAA and 12 articles examining the shared risk factor hypothesis were identified.There is insufficient evidence to confirm that AAAs are the result of advanced atherosclerosis. The bulk of evidence points to AAA disease being a systemic disease of the vasculature, with a predetermined genetic susceptibility leading to a phenotype governed by environmental factors.

Current evidence suggests that carotid endarterectomy (CEA) performed within 2 weeks of symptoms produces better long-term results than if it is delayed. Urgent endarterectomy following unstable presentations such as crescendo transient ischemic attack (cTIA) or progressive stroke has been associated with variable results. The evidence for this treatment strategy required reviewing. A systematic review of articles related to urgent CEA between 1980 and 2008 was performed. For cTIA, there was an odds ratio of 5.6 (95% confidence interval 3.3-9.7, p < or = .0001) for combined stroke or death compared with surgery for "standard" indications. For unstable stroke, the odds ratio was 5.5 (95% confidence interval 3.1-9.3, p < or = .0001). Patients with unstable neurologic presentations are at higher risk of complications if operated on urgently. Clearer definitions would help more precise patient selection to avoid inadvertently operating on patients with an unacceptably high risk of poor outcome.

Post hemorrhagic hydrocephalus (PHH) can be defined as progressive dilation of the ventricular system that develops as a complication of neonatal intraventricular hemorrhage (IVH). Grading systems exist to quantify IVH but not this secondary ventricular dilation. Cranial ultrasound techniques and measuring methods that allow a uniform, objective grading convention for ventricular enlargement are presented. A biventricular to biparietal ratio (BV:BD) can be measured in the coronal images. A ventricular diameter (VD) can be measured in the sagittal images. Cortical mantle thickness and ventricular size can be measured in the axial images. These measurements allow more precise identification of changes in ventricular size which may have treatment implications. The clinical features of PHH and treatment options are also reviewed. No one treatment modality has proven superior, and neurologic outcome does not correlate with development or resolution of PHH. Better description of parenchymal damage is needed during routine ultrasound studies and development of other modalities to assess parenchymal function need to be developed before the significance of IVH and PHH can be fully understood.