Cumulative sum analysis score and phacoemulsification competency learning curve

AIM： To use the cumulative sum analysis score（CUSUM） to construct objectively the learning curve of phacoemulsification competency.METHODS： Three second-year residents and an experienced consultant were monitored for a series of 70 phacoemulsification cases each and had their series analysed by CUSUM regarding posterior capsule rupture（PCR） and best-corrected visual acuity. The acceptable rate for PCR was 〈5%（lower limit h） and the unacceptable rate was 〉10%（upper limit h）. The acceptable rate for bestcorrected visual acuity worse than 20/40 was 〈10%（lower limit h） and the unacceptable rate was 〉20%（upper limit h）. The area between lower limit h and upper limit h is called the decision interval. RESULTS： There was no statistically significant difference in the mean age, sex or cataract grades between groups. The first trainee achieved PCR CUSUM competency at his 22 nd case. His best-corrected visual acuity CUSUM was in the decision interval from his third case and stayed there until the end, never reaching competency. The second trainee achieved PCR CUSUM competency at his 39^ th case. He could reach best-corrected visual acuity CUSUM competency at his 22 ^nd case. The third trainee achieved PCR CUSUM competency at his 41 st case. He reached bestcorrected visual acuity CUSUM competency at his 14 ^th case.CONCLUSION： The learning curve of competency in phacoemulsification is constructed by CUSUM and in average took 38 cases for each trainee to achieve it.

Using junction trees for structural learning of Bayesian networks

The learning Bayesian network （BN） structure from data is an NP-hard problem and still one of the most exciting chal- lenges in the machine learning. In this work, a novel algorithm is presented which combines ideas from local learning, constraint- based, and search-and-score techniques in a principled and ef- fective way. It first reconstructs the junction tree of a BN and then performs a K2-scoring greedy search to orientate the local edges in the cliques of junction tree. Theoretical and experimental results show the proposed algorithm is capable of handling networks with a large number of variables. Its comparison with the well-known K2 algorithm is also presented.

Synergistic application of molecular docking and machine learning for improved binding pose

Accurate prediction of protein-ligand complex structures is a crucial step in structure-based drug design.Traditional molecular docking methods exhibit limitations in terms of accuracy and sampling space,while relying on machine-learning approaches may lead to invalid conformations.In this study,we propose a novel strategy that combines molecular docking and machine learning methods.Firstly,the protein-ligand binding poses are predicted using a deep learning model.Subsequently,position-restricted docking on predicted binding poses is performed using Uni-Dock,generating physically constrained and valid binding poses.Finally,the binding poses are re-scored and ranked using machine learning scoring functions.This strategy harnesses the predictive power of machine learning and the physical constraints advantage of molecular docking.Evaluation experiments on multiple datasets demonstrate that,compared to using molecular docking or machine learning methods alone,our proposed strategy can significantly improve the success rate and accuracy of protein-ligand complex structure predictions.