人工智能在中医药领域的应用进展及现状思考

人工智能（Artificial Intelligence，AI）的蓬勃兴起为现代社会带来了前所未有的机遇，中医药是中华民族传承千年的文化瑰宝。随着人工智能技术不断在中医药领域的科技创新中崭露头角，二者的融合不断加深，人工智能在中医药领域的发展前景、争议挑战也引发了诸多思考。本文将从人工智能在中医药领域的应用入手，对人工智能辅助中医诊断、智能决策与数据挖掘、健康管理及中草药现代化研究等方面，就近年来国内外研究进展进行总结与分析，以期为AI视域下实现中医药现代化、智能化赋能。     

Prior surgical fixation of proximal humerus fractures is associated with increased complications following subsequent reverse shoulder arthroplasty

BackgroundThe objective of this study was to compare complication rates between patients undergoing reverse shoulder arthroplasty (RSA) after a prior open reduction and internal fixation (ORIF) for proximal humerus fracture (PHF) to those undergoing RSA as a primary treatment for PHFs, glenohumeral osteoarthritis, or rotator cuff tear arthropathy (CTA).MethodsPatients who underwent RSA between 2015 and 2020 were identified in the Mariner database. Patients were separated into 3 mutually exclusive groups: (1) RSA for osteoarthritis, rotator cuff tear, or CTA (Control-RSA); (2) RSA as a primary treatment for PHF (PHF-RSA); and (3) RSA for patients with prior ORIF of PHFs (ORIF-RSA). Ninety-day medical and 2-year postoperative surgical complications were identified. In addition, patients in the PHF-RSA group were subdivided into those undergoing RSA for PHF within 3 months of the fracture (acute) vs. those treated greater than 3 months from diagnosis (delayed). Multivariate regression was performed to control for differences in comorbidities and demographics.ResultsA total of 30,824 patients underwent primary RSA for arthritis or CTA, 5389 patients underwent RSA as a primary treatment for a PHF, and 361 patients underwent RSA after ORIF of a PHF. ORIF before RSA was associated with an increased risk of overall revision (odds ratio [OR] 2.45, P = .002), infection (OR 2.40, P < .001), instability (OR 2.43, P < .001), fracture (OR 3.24, P = .001), minor medical complications (OR 1.59, P = .008), and readmission (OR 2.55, P = .001) compared with the Control-RSA cohort. RSA as a primary treatment for PHF was associated with an increased risk of 2-year revision (OR 1.60, P < .001), infection (OR 1.51, P < .001), instability (OR 2.84, P < .001), and fracture (OR 2.54, P < .001) in addition to major medical complications (OR 2.02, P < .001), minor medical complications (OR 1.92, P < .001), 90-day emergency department visits (OR 1.26, P < .001) and 90-day readmission (OR 2.03, P < .001) compared with the Control-RSA cohort. The ORIF-RSA group had an increased risk of periprosthetic infection (OR 1.94, P = .002) when compared with the PHF-RSA cohort. There were no differences in medical or surgical complications in the RSA-PHF cohort between patients treated in an acute or delayed fashion.ConclusionRSA following ORIF of a PHF is associated with increased complications compared with patients undergoing RSA for nonfracture indications. Prior ORIF of a PHF is also an independent risk factor for postoperative infection after RSA compared with patients who undergo RSA as a primary operation for fracture. The timing of RSA as a primary operation for PHF does not appear to impact the rates of postoperative medical and surgical complications.     

Dry eye signs and symptoms in aromatase inhibitor treatment and the relationship with pain

PurposeAromatase inhibitors (AIs) limit the synthesis of oestrogen in peripheral tissues thus lowering levels of oestrogen. The primary aim was to evaluate whether women treated with AIs have altered dry eye symptoms and signs. A sub-aim was to investigate whether symptoms of dry eye in postmenopausal women were associated with symptoms of non-eye pain, ocular pain and self-rated pain perception.MethodsThis cross-sectional, observational, single visit study recruited 56 postmenopausal women (mean age 64.1 + 7.9 years) and 52 undergoing AI treatment (mean age 66.6 + 9.0). Ocular symptoms (OSDI, MGD14) and pain questionnaires (PSQ, OPAS) were administered and signs of dry eye and meibomian gland dysfunction were evaluated.ResultsAlmost half of each group reported dry eye symptoms, defined as OSDI>12 (48% control, 46% AI). The PSQ score was significantly higher in the AI group (p = 0.04). Neither frequency or severity of dry eye (or MGD) symptoms scores were significantly different between groups. In the AI group, meibomian gland expressibility score was worse (p = 0.003); there were no differences in any other signs. Higher OSDI scores were associated with higher OPAS eye-pain scores (r = 0.49, p < 0.001), but not OPAS non-eye pain (r = 0.09, p = 0.35). Pain perception (PSQ) showed a moderate positive association with OPAS eye-pain (r = 0.30, p = 0.003).ConclusionsIn this study elevated ocular symptoms were observed in both the AI treated and the untreated groups, with no difference between the groups. Women undergoing AI treatment for early stage breast cancer had worse meibum expressibility score and increased pain perception compared to an untreated group of women.     

后外侧结构重建在新鲜股骨颈骨折人工股骨头置换术中的应用

目的探讨后外侧结构重建对后外侧入路人工股骨头置换术术后早期关节脱位的影响。方法选取2016年9月至2017年8月于我院行后外侧入路初次人工股骨头置换术的股骨颈骨折患者60例,根据术中是否修补关节囊及外旋肌群分为重建组(33例:舌形切开关节囊,术中将关节囊及外旋肌群原位缝合在大转子后方及臀中肌肌腱附着处)和对照组(27例:切除关节囊后,术中未进行外旋肌群修复重建)。比较两组的手术情况及术后近期关节功能情况。结果重建组的手术时间为(45.0±15.3) min,长于对照组的(35.0±12.4) min (P <0.05)。重建组术腔引流量为(200.0±80.0) m L,少于对照组的(420.0±120.6) m L (P <0.05)。重建组的早期脱位率为0.000%(0例),与对照组的7.407%(2例)比较无统计学差异(P>0.05)。重建组术后Harris评分为(92.0±3.4)分,高于对照组的(88.2±5.0)分(P <0.05)。结论在后外侧入路人工股骨头置换过程中行后外侧结构重建能够有效减少术腔引流量,提高髋关节Harris评分,对维持髋关节软组织平衡具有重要意义。     

Machine learning models for Parkinson’s disease detection and stage classification based on spatial-temporal gait parameters

BackgroundParkinson’s disease (PD) is a chronic and progressive neurodegenerative disease with no cure, presenting a challenging diagnosis and management. However, despite a significant number of criteria and guidelines have been proposed to improve the diagnosis of PD and to determine the PD stage, the gold standard for diagnosis and symptoms monitoring of PD is still mainly based on clinical evaluation, which includes several subjective factors. The use of machine learning (ML) algorithms in spatial-temporal gait parameters is an interesting advance with easy interpretation and objective factors that may assist in PD diagnostic and follow up.Research questionThis article studies ML algorithms for: i) distinguish people with PD vs. matched-healthy individuals; and ii) to discriminate PD stages, based on selected spatial-temporal parameters, including variability and asymmetry.MethodsGait data acquired from 63 people with PD with different levels of PD motor symptoms severity, and 63 matched-control group individuals, during self-selected walking speed, was study in the experiments.ResultsIn the PD diagnosis, a classification accuracy of 84.6 %, with a precision of 0.923 and a recall of 0.800, was achieved by the Naïve Bayes algorithm. We found four significant gait features in PD diagnosis: step length, velocity and width, and step width variability. As to the PD stage identification, the Random Forest outperformed the other studied ML algorithms, by reaching an Area Under the ROC curve of 0.786. We found two relevant gait features in identifying the PD stage: stride width variability and step double support time variability.SignificanceThe results showed that the studied ML algorithms have potential both to PD diagnosis and stage identification by analysing gait parameters.     

Registries,Databases and Repositories for Developing Artificial Intelligence in Cancer Care

Modern artificial intelligence techniques have solved some previously intractable problems and produced impressive results in selected medical domains. One of their drawbacks is that they often need very large amounts of data. Pre-existing datasets in the form of national cancer registries, image/genetic depositories and clinical datasets already exist and have been used for research. In theory, the combination of healthcare Big Data with modern, data-hungry artificial intelligence techniques should offer significant opportunities for artificial intelligence development, but this has not yet happened. Here we discuss some of the structural reasons for this, barriers preventing artificial intelligence from making full use of existing datasets, and make suggestions as to enable progress. To do this, we use the framework of the 6Vs of Big Data and the FAIR criteria for data sharing and availability (Findability, Accessibility, Interoperability, and Reuse). We share our experience in navigating these barriers through The Brain Tumour Data Accelerator, a Brain Tumour Charity-supported initiative to integrate fragmented patient data into an enriched dataset. We conclude with some comments as to the limits of such approaches.