Learning curve

The paper presents an efficient 88 line MATLAB code for topology optimization. It has been developed using the 99 line code presented by Sigmund (Struct Multidisc Optim 21(2):120–127, 2001) as a starting point. The original code has been extended by a density filter, and a considerable improvement in efficiency has been achieved, mainly by preallocating arrays and vectorizing loops. A speed improvement with a factor of 100 is obtained for a benchmark example with 7,500 elements. Moreover, the length of the code has been reduced to a mere 88 lines. These improvements have been accomplished without sacrificing the readability of the code. The 88 line code can therefore be considered as a valuable successor to the 99 line code, providing a practical instrument that may help to ease the learning curve for those entering the field of topology optimization. The paper also discusses simple extensions of the basic code to include recent PDE-based and black-and-white projection filtering methods. The complete 88 line code is included as an appendix and can be downloaded from the web site www.topopt.dtu.dk.

Background Given the increasing demand for tissue-sparing surgery, the surgical approach is the subject of lively debate in total hip replacement. The aim of this paper is to compare the efficacy of the minimally invasive direct anterior approach and the standard lateral approach to total hip replacement surgery by observing intra- and perioperative outcomes. Materials and methods The authors conducted a retrospective study on a group of 419 consecutive patients undergoing total hip replacement for coxarthrosis. The patients were divided into a first group (A) of 198 patients who had surgery with the standard lateral approach, and a second control group (B) of 221 patients who had the same procedure via the minimally invasive direct anterior approach. Assessment of the two groups considered the following perioperative parameters: length of the surgical procedure, intraoperative complications, intra- and postoperative blood loss, postoperative pain, postoperative nausea and vomiting, length of stay, and type of discharge. Results The two groups were homogeneous when compared in relation to mean age, sex and body weight. The minimally invasive direct anterior approach was performed within an acceptable time (89 ± 19 min vs. 81 ± 15 min) and with modest blood loss (3.1 ± 0.9 g/dL vs. 3,5 ± 1 g/dL). Patients experienced less pain (1.4 ± 1.5 NRS score vs. 2.5 ± 2 NRS score), and PONV affected only 5% versus 10% of cases. Times to discharge were shorter (7 ± 2 days vs. 10 ± 3.5 days), and 58.4% versus 11.6% of patients were discharged to home. Conclusions In our study, patients treated with a minimally invasive direct anterior approach had a better perioperative outcome than patients treated with the lateral approach. The longer time of surgery for the minimally invasive direct anterior approach may be attributed to the learning curve. Further studies are necessary to investigate the advantages of a minimally invasive direct anterior approach in terms of clinical results in the short and long run.

In the context of classification problems, algorithms that generate multivariate trees are able to explore multiple representation languages by using decision tests based on a combination of attributes. In the regression setting, model trees algorithms explore multiple representation languages but using linear models at leaf nodes. In this work we study the effects of using combinations of attributes at decision nodes, leaf nodes, or both nodes and leaves in regression and classification tree learning. In order to study the use of functional nodes at different places and for different types of modeling, we introduce a simple unifying framework for multivariate tree learning. This framework combines a univariate decision tree with a linear function by means of constructive induction. Decision trees derived from the framework are able to use decision nodes with multivariate tests, and leaf nodes that make predictions using linear functions. Multivariate decision nodes are built when growing the tree, while functional leaves are built when pruning the tree. We experimentally evaluate a univariate tree, a multivariate tree using linear combinations at inner and leaf nodes, and two simplified versions restricting linear combinations to inner nodes and leaves. The experimental evaluation shows that all functional trees variants exhibit similar performance, with advantages in different datasets. In this study there is a marginal advantage of the full model. These results lead us to study the role of functional leaves and nodes. We use the bias-variance decomposition of the error, cluster analysis, and learning curves as tools for analysis. We observe that in the datasets under study and for classification and regression, the use of multivariate decision nodes has more impact in the bias component of the error, while the use of multivariate decision leaves has more impact in the variance component.

When participants studied a list of paired associates for several study-test cycles, their judgments of learning (JOLs) exhibited relatively good calibration on the 1st cycle, with a slight overconfidence. However, a shift toward marked underconfidence occurred from the 2nd cycle on. This underconfidence-with-practice (UWP) effect was very robust across several experimental manipulations, such as feedback or no feedback regarding the correctness of the answer, self-paced versus fixed-rate presentation, different incentives for correct performance, magnitude and direction of associative relationships, and conditions producing different degrees of knowing. It was also observed both in item-by-item JOLs and in aggregate JOLs. The UWP effect also occurred for list learning and for the memory of action events. Several theoretical explanations for this counterintuitive effect are discussed.

Background: After more than 8 years of working in the field, we thought it would be interesting to evaluate our experience in the laparoscopic repair of abdominal wall hernias, focusing attention on the lessons learned with time. Methods: From January 1994 to November of 2000, a total of 270 patients with abdominal wall hernias were treated in our center using the laparoscopic approach. The data collected and analyzed were preoperative evaluation, operative findings, early and long-term complications, and recurrences. Results: The mean follow-up time was 44 months, mean surgical time was 85 min, and mean hospital stay was 1.5 days. The average number of abdominal wall defects was 4.8 per patient. There were 9 (3.3%) small bowel perforations. Conversion to open surgery was required in 1 case (0.3%). Minor early postoperative complications occurred in 38 patients (14.07%). Twenty patients (7.4%) developed persistent postoperative abdominal pain. There was 1 case (0.3%) of small bowel incarceration through the mesh border and another case (0.3%) of small bowel leakage due to ischemia and subsequent peritonitis. The relapse rate was 4.4% (12 cases). Conclusion: The laparoscopic approach is a valuable option in the management of abdominal wall hernias, but it requires experience in laparoscopic surgery and there is a specific learning curve for the technique.

Background: Lymphatic mapping and sentinel lymph node (SLN) biopsy are new techniques that accurately provide crucial staging information while inflicting far less morbidity than complete axillary dissection. As these techniques continue to gain acceptance, issues such as adequacy of training, certification, and outcomes measures become increasingly important. The purpose of this paper is to report the initial lymphatic mapping experience at the H Lee Moffitt Cancer Center and Research Institute and to provide a detailed description of the technical aspects of lymphatic mapping.Study Design: From April 1994 to April 1998, 700 patients with newly diagnosed breast cancers underwent an IRB-approved prospective trial of lymphatic mapping using a combination of Lymphazurin (USSC, Norwalk, CT) blue dye and filtered technetium 99m-labeled sulfur-colloid. Failure of the procedure was defined as the inability to detect an SLN by either radiocolloid uptake within a lymph node by the gamma probe or the inability to visualize blue staining of a lymph node. Learning curves were then generated as the failure rate versus serial number of patients for each of the 5 surgeons involved in this study.Results: The SLN was identified in 665 of 700 patients (95.0%). A total of 1,348 SLNs were successfully removed, of which 238 (17.7%) were positive for metastatic disease in 176 of 665 patients (26.5%). In patients who underwent a complete axillary dissection after SLN biopsy, SLNs were identified in 173 of 186 patients (93.0%). Of the 173 patients, 53 patients (30.6%) had positive SLNs and 120 patients (69.4%) had negative SLNs. In the 120 patients with negative SLNs, one patient was found to have disease on complete dissection, for a false-negative rate of 0.83% (95% CI: 0.02%, 4.6%). A learning curve representing the mean of the 5 surgeons’ experience indicates that on average 23 patients are required by an individual surgeon to achieve a 90% ± 4.5% success rate and 53 patients are required to achieve a 95% ± 2.3% success rate (p = 0.05).Conclusions: These data validate lymphatic mapping and SLN biopsy as indispensable tools in the surgical treatment of breast cancer. With adequate multidisciplinary training, these techniques can be readily implemented at institutions treating breast cancer.

Background: Robotically enhanced telemanipulation for totally endoscopic coronary artery bypass does not provide adequate tactile feedback, traction, or countertraction. The exposition of coronary target sites is difficult, the visual field is limited, and the epicardial stabilization may be troublesome. A fourth robotic arm for endothoracic instrumentation has been added to the da Vinci surgical system to facilitate totally endoscopic operations. The stereoendoscope was upgraded with a wide-angle feature. Methods: The procedure was performed in five patients. Four of these patients had left internal thoracic artery (LITA) to left anterior descending artery (LAD) grafting on the beating heart and the fifth had sequential bypass grafting (LITA to diagonal branch and LAD) on an arrested heart. The additional effector arm of the da Vinci surgical system was brought into the operative field beneath the operating table and used as a second right arm. The wide-angle view was activated by either the console or the patient side surgeon. Results: The mean operative, port placement, and anastomotic times for a beating-heart totally endoscopic coronary artery bypass were 195 ± 58, 25 ± 10, and 18 ± 5 min, respectively. All procedures were free of morbidity and mortality, with satisfactory angiographic control. The sequential arterial bypass grafting procedure was fully completed in totally endoscopic technique. Conclusions: The additional instrumentation arm and wide-angle visualization are useful technical improvements of the da Vinci surgical system, solving the problem of traction, countertraction, and facilitated exposition of target sites as well as visualization of the surgical field. They provide potential for wider acceptance of totally endoscopic coronary artery bypass grafting in a larger surgical community.

Background Robotic telemanipulation systems provide solutions to the problems of less dexterity and visual constraints of minimally invasive surgery (MIS). However, their influence over surgeons’ dexterity and learning curve needs to be assessed. We present motion analysis as an objective method to measure performance and learning progress. Methods Thirteen surgeons completed five synthetic small bowel anastomoses using the da Vinci system. Objective Structured Assessment of Technical Skills (OSATS) allowed qualitative analysis. Quantitative analysis used API software of the system to retrieve real-time robotic signal data of time, path length, and number of movements. Wilcoxon signed ranks test was used for statistical analysis. A p value <0.05 was considered significant. Results OSATS global scores were 18.6 points for the first attempt and 26 for the fifth attempt (p < 0.02, Cronbach’s alpha = 0.894). Paired data of motion analysis for attempts 1 vs 5 showed significant change: time taken 3507 sec and 2287 sec (p < 0.008), total number of movements 2411 and 1387 (p = 0.01), total path length 21,630 cm and 13,941 cm (p = 0.01). Conclusions A rapid learning curve to a competent level using the da Vinci system is possible aided by the system’s intuitive motion. Motion analysis is a useful tool to measure performance in the da Vinci system compared to OSATS and time alone.