A Curriculum Batching Strategy for Automatic ICD Coding with Deep Multi-Label Classification Models.

The International Classification of Diseases (ICD) has an important role in building applications for clinical medicine. Extremely large ICD coding label sets and imbalanced label distribution bring the problem of inconsistency between the local batch data distribution and the global training data distribution into the minibatch gradient descent (MBGD)-based training procedure for deep multi-label classification models for automatic ICD coding. The problem further leads to an overfitting issue.

[Problems and solutions on transformation of scientific research achievements of acupuncture].

With more and more attention and investment on acupuncture scientific researches, considerable outcomes and achievements has been acquired, but the shortcoming of low transformation rate of acupuncture research achievements is gradually exposed. Nowadays there is no related report on this problem, so based on achievement translational research in other areas and practical situation of acupuncture, the existing problems and solutions are analyzed. As a result, the existing problems include (1) the research content is mainly basic research and clinical research but less acupuncture device research, leading to limited transformation efficiency; (2) the evaluation system and transformation pattern are still needed to be perfect.

[Our considerations about research on regularities of acupoint combination based on bipartite network community structure partition].

Acupoint compatibility is not only a foundation for formulating acupoint recipes in clinical practice, but also a core of meridian and acupoint theory. Many studies have revealed that the application of multiple-acupoints combination in the treatment of different clinical conditions or illnesses may lead to three outcomes, i.e.