Language Representation Models: An Overview.

In the last few decades, text mining has been used to extract knowledge from free texts. Applying neural networks and deep learning to natural language processing (NLP) tasks has led to many accomplishments for real-world language problems over the years. The developments of the last five years have resulted in techniques that have allowed for the practical application of transfer learning in NLP. The advances in the field have been substantial, and the milestone of outperforming human baseline performance based on the general language understanding evaluation has been achieved. This paper implements a targeted literature review to outline, describe, explain, and put into context the crucial techniques that helped achieve this milestone. The research presented here is a targeted review of neural language models that present vital steps towards a general language representation model.

Retroperitoneal Versus Transperitoneal Robotic Partial Nephrectomy: A Multicenter Matched-pair Analysis.

BACKGROUND: With increasing acceptance of robotic partial nephrectomy over the last decade, there is an ongoing discussion about the transperitoneal versus retroperitoneal access. OBJECTIVE: To report outcomes after transperitoneal (TRPN) versus retroperitoneal (RRPN) robotic partial nephrectomy. DESIGN, SETTING, AND PARTICIPANTS: A total of 754 patients were identified from the databases of three high-volume centers who underwent either TRPN (n = 551) or RRPN (n = 203). OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: Perioperative data were evaluated retrospectively. A propensity score matched-pair analysis was performed with the following variables: grade of renal insufficiency, age, body mass index, tumor diameter, and preoperative aspects and dimensions used for an anatomical (PADUA) score with a subsequent subgroup analysis for tumor location. For quality outcomes, the margin, ischemia, and complications (MIC) criteria were used (negative margins, ischemia time <20 min, and no major complications). Statistical analyses included chi-square and Mann-Whitney U tests. RESULTS AND LIMITATIONS: In all, 176 patients could be matched in each group. The median tumor diameter was 28 mm with a PADUA score of 9. In 11% of RRPN versus 44% of TRPN cases, an anterior tumor location was found, and in 55% versus 30%, a posterior lesion was found (p < 0.001). Operative time (119 vs 139 min, p < 0.001) and warm ischemia time (9 vs 10 min, p = 0.003) were significantly shorter for RRPN. No significant differences were observed between intra- and postoperative complication rates, with 8% major complications in TRPN versus 3% in RRPN (p = 0.06). The MIC criteria were achieved in 90% in the RRPN versus 88% in the TRPN group, without differences for tumor location. CONCLUSIONS: Significant differences between TRPN and RRPN could be found for intraoperative time, while complication rates and quality outcomes were comparable. RRPN can also be a considerable alternative for anterior tumors. PATIENT SUMMARY: In this study, we demonstrate that robotic partial nephrectomy is feasible with either a transperitoneal or a retroperitoneal surgical access. The posterior approach can also be used for anterior renal tumors and may result in shorter operative time.

Robot-assisted partial nephrectomy: A single-center matched-pair analysis of the retroperitoneal versus the transperitoneal approach.

OBJECTIVE: Comparison of the retroperitoneal (RRPN) perioperative variables and the transperitoneal (TRPN) robot-assisted partial nephrectomy (RPN) using a matched-pair analysis. MATERIAL AND METHODS: A retrospective review was carried out for 224 patients who underwent RPN between 2014 and 2019. A matched-pair analysis was performed on 51 pairs of patients. The matching criteria were age, Charlson comorbidity index, body mass index, the grade of renal insufficiency, tumor diameter, and Preoperative Aspects and Dimensions Used for an Anatomical Classification of Renal Tumors score. RESULTS: The time to reach the renal hilum (P < .001), the overall complication rate (P » .008), and the major complication rate (P » .01) were lower in the RRPN group. The operative time was 143 vs 150minutes (P » .63) in RRPN vs TRPN, respectively. Warm ischemia time was 10minutes in RRPN vs 12minutes in TRPN (P » .07). Early unclamping was used in 71% in RRPN vs 48% in TRPN (P » .02). The length of hospital stay was 6 days in both groups (P » .11). The cases' complexity, the rate of positive surgical margins, and postoperative kidney function were comparable in both groups (P > .05). CONCLUSION: The advantages of RRPN lie in the shorter time to reach the renal hilum and the lower complication rates; the comparability with the other parameters proves the safety and feasibility of the RRPN access for localized kidney tumors.