Neural specificity of acupuncture stimulation from support vector machine classification analysis

The Use of Artificial Intelligence in Complementary and Alternative Medicine: A Systematic Scoping Review

Background: The development of artificial intelligence (AI) in the medical field has been growing rapidly. As AI models have been introduced in complementary and alternative medicine (CAM), a systematized review must be performed to understand its current status.

Objective: To categorize and seek the current usage of AI in CAM.

Method: A systematic scoping review was conducted based on the method proposed by the Joanna Briggs Institute. The three databases, PubMed, Embase, and Cochrane Library, were used to find studies regarding AI and CAM. Only English studies from 2000 were included. Studies without mentioning either AI techniques or CAM modalities were excluded along with the non-peer-reviewed studies. A broad-range search strategy was applied to locate all relevant studies.

Results: A total of 32 studies were identified, and three main categories were revealed: 1) acupuncture treatment, 2) tongue and lip diagnoses, and 3) herbal medicine. Other CAM modalities were music therapy, meditation, pulse diagnosis, and TCM syndromes. The majority of the studies utilized AI models to predict certain patterns and find reliable computerized models to assist physicians.

Conclusion: Although the results from this review have shown the potential use of AI models in CAM, future research ought to focus on verifying and validating the models by performing a large-scale clinical trial to better promote AI in CAM in the era of digital health.

Machine Learning in Neuroimaging: A New Approach to Understand Acupuncture for Neuroplasticity

The effects of acupuncture facilitating neural plasticity for treating diseases have been identified by clinical and experimental studies. In the last two decades, the application of neuroimaging techniques in acupuncture research provided visualized evidence for acupuncture promoting neuroplasticity. Recently, the integration of machine learning (ML) and neuroimaging techniques becomes a focus in neuroscience and brings a new and promising approach to understand the facilitation of acupuncture on neuroplasticity at the individual level. This review is aimed at providing an overview of this rapidly growing field by introducing the commonly used ML algorithms in neuroimaging studies briefly and analyzing the characteristics of the acupuncture studies based on ML and neuroimaging, so as to provide references for future research.

Does the Acupoint Specificity Exist? Evidence from Functional Neuroimaging Studies

BACKGROUND

Using functional neuroimaging techniques to explore the central mechanism of acupoint specificity, the key of acupuncture theory and clinical practice, has attracted increasing attention worldwide. This review aimed to investigate the current status of functional neuroimaging studies on acupoint specificity and explore the potential influencing factors for the expression of acupoint specificity in neuroimaging studies.

METHODS

PubMed database was searched from January 1st, 1995 to December 31st, 2016 with the language restriction in English. Data including basic information, methodology and study results were extracted and analyzed from the eligible records.

RESULTS

Seventy-nine studies were finally enrolled. 65.8% of studies were performed in China, 73.4% of studies were conducted with healthy subjects, 77.2% of studies chose manual acupuncture as the intervention, 86.1% of studies focused on the instant efficacy and 89.9% of studies used functional magnetic resonance imaging as scanning technique. The average sample size was 16 per group. The comparison of verum acupoints and sham acupoints were the main body of acupoint specificity researches. 93.7% of studies obtained the positive results and favored the existence of acupoint specificity.

CONCLUSION

This review affirmed the existence of acupoint specificity and deemed that the acupoint specificity was relative. Multiple factors such as participants, sample size, acupoint combinations, treatment courses, and types of acupoint could influence the expression of acupoint specificity.

Effect of Compound Laser Acupuncture-Moxibustion on Blood Glucose, Fasting Insulin and Blood Lipids Levels in Type 2 Diabetic Rats

OBJECTIVE

To investigate the effect of compound laser acupuncture-moxibustion on blood glucose, fasting insulin and blood lipids levels in type 2 diabetes mellitus (T2DM) rats.

METHODS

Forty male Wistar rats were randomly divided into 4 groups, including the normal group, model control group, laser group and sham laser group (n=10 per group). The rats in the normal group were fed with a standard diet. Rats in other groups were fed with a high-sugar and high-fat diet for 4 weeks, then intraperitoneally injected with 1% streptozotocin to induce T2DM model. The laser group was irradiated by 10.6 µm and 650 nm compound laser on bilateral Pishu (BL 20), Shenshu (BL 23) and Sanyinjiao (SP 6) for 5 min, 6 times a week for 5 weeks. The sham laser group received the same treatment as the laser group, but without laser output. The model control group and normal group were not treated. Blood glucose levels were measured before and after 1, 2, 3, 4 and 5 weeks of treatment. The serum levels of fasting insulin, total cholesterol (TC), triglyceride (TG), low-density lipoprotein (LDL), and high-density lipoprotein (HDL) were analyzed after the last treatment.

RESULTS

The blood glucose levels in the model control group increased during the 5 weeks of treatment compared with the normal group (P<0.05), while those in the laser group were significantly lower than the model control group after weekly treatment (P<0.01 or P<0.05). After 1, 2 and 3 weeks of treatment, the blood glucose levels in the laser group decreased obviously compared with the sham laser group (P<0.01 or P<0.05). Compared with the normal group, the levels of fasting insulin, TC and LDL in the model control group notably increased (P<0.01 or P<0.05), while their levels in the laser group were significantly lower than the model control group after 5 weeks of treatment (P<0.05 or P<0.01). However, no statistically significant differences were observed in TG or HDL levels among the 4 groups (P>0.05).

CONCLUSION

The compound laser acupuncture-moxibustion of 10.6 µm and 650 nm had positive effects on the regulation of hyperglycemia and insulin resistance in T2DM rats, which may be a potential treatment for T2DM, and also provide an alternative to the traditional acupuncture and moxibustion therapy.

Modulation Effect of Acupuncture on Functional Brain Networks and Classification of Its Manipulation With EEG Signals

Acupuncture manipulation is the key of Chinese medicine acupuncture therapy. In clinical practice, different acupuncture manipulations are required to achieve different therapeutic effects, which means it is crucial to distinguish different acupuncture manipulations. In this paper, we proposed a classification framework for different acupuncture manipulations, which employed the graph theory and machine learning method. Multichannel EEG signals evoked by acupuncture at "Zusanli" acupoint were recorded from healthy humans by two acupuncture manipulations: twirling-rotating (TR) and lifting-thrusting (LT). Phase locking value was used to estimate the phase synchronization of pair-wise EEG channels. It was found that acupunctured by TR manipulation exhibit significantly higher synchronization degree than acupunctured by LT manipulation. With the construction of functional brain network, the topological features of graph theory were extracted. Taken the network features as inputs, machine learning classifiers were established to classify acupuncture manipulations. The highest accuracy can achieve 92.14% with support vector machine. By further optimizing the network features utilized in machine learning classifiers, it was found that the combination of node betweenness and small world network index is the most effective factor for acupuncture manipulations classification. These findings suggested that our approach provides new ideas for automatically identify acupuncture manipulations from the perspective of functional brain networks and machine learning methods.

Identification of Alzheimer’s Disease on the Basis of a Voxel-Wise Approach

Robust prediction of Alzheimer’s disease (AD) helps in the early diagnosis of AD and may support the treatment of AD patients. In this study, for early detection of AD and prediction of mild cognitive impairment (MCI) conversion, we develop an automatic computer-aided diagnosis (CAD) framework based on a merit-based feature selection method through a whole-brain voxel-wise analysis using baseline magnetic resonance imaging (MRI) data. We also explore the impact of different MRI spatial resolution on the voxel-wise metric AD classification and MCI conversion prediction. We assessed the proposed CAD framework using the whole-brain voxel-wise MRI features of 507 J-ADNI participants (146 healthy controls [HCs], 102 individuals with stable MCI [sMCI], 112 with progressive MCI [pMCI], and 147 with AD) among four clinically relevant pairs of diagnostic groups at different imaging resolutions (i.e., 2, 4, 8, and 16 mm). Using a support vector machine classifier through a 10-fold cross-validation strategy at a spatial resolution of 2 mm, the proposed CAD framework yielded classification accuracies of 91.13%, 74.77%, 81.12%, and 81.78% in identifying AD/healthy control, sMCI/pMCI, sMCI/AD, and pMCI/HC, respectively. The experimental results show that a lower spatial resolution (i.e., 2 mm) may provide more robust information to trace the neuronal loss-related brain atrophy in AD.

Classification of Alzheimer's Disease and Prediction of Mild Cognitive Impairment Conversion Using Histogram-Based Analysis of Patient-Specific Anatomical Brain Connectivity Networks

In this study, we investigated the early detection of Alzheimer's disease (AD) and mild cognitive impairment (MCI) conversion to AD through individual structural connectivity networks using structural magnetic resonance imaging (sMRI) data. In the proposed method, the cortical morphometry of individual gray matter images were used to construct structural connectivity networks. A statistical feature generation approach based on histogram-based feature generation procedure was proposed to represent a statistical-pattern of connectivity networks from a high-dimensional space into low-dimensional feature vectors. The proposed method was evaluated on numerous samples including 61 healthy controls (HC), 42 stable-MCI (sMCI), 45 progressive-MCI (pMCI), and 83 AD subjects at the baseline from the J-ADNI data-set using support vector machine classifier. The proposed method yielded a classification accuracy of 84.17%, 70.38%, and 61.05% in identifying AD/HC, MCIs/HCs, and sMCI/pMCI, respectively. The experimental results show that the proposed method performed in a comparable way to alternative methods using MRI data.

Brain Connectivity Patterns Dissociate Action of Specific Acupressure Treatments in Fatigued Breast Cancer Survivors

Persistent fatigue is a pernicious symptom in many cancer survivors. Existing treatments are limited or ineffective and often lack any underlying biologic rationale. Acupressure is emerging as a promising new intervention for persistent cancer-related fatigue; however, the underlying mechanisms of action are unknown. Our previous investigations suggested that fatigued breast cancer survivors have alterations in brain neurochemistry within the posterior insula and disturbed functional connectivity to the default mode network (DMN), as compared to non-fatigued breast cancer survivors. Here, we investigated if insula and DMN connectivity were modulated by self-administered acupressure by randomizing breast cancer survivors (n = 19) to two distinct treatments: relaxing acupressure or stimulating acupressure. All participants underwent proton magnetic resonance spectroscopy of the posterior insula and functional connectivity magnetic resonance imaging at baseline and immediately following 6 weeks of acupressure self-treatment. As compared to baseline measures, relaxing acupressure decreased posterior insula to dorsolateral prefrontal cortex connectivity, whereas stimulating acupressure enhanced this connectivity (p < 0.05 corrected). For relaxing but not stimulating acupressure, reduced connectivity was associated with sleep improvement. In addition, connectivity of the DMN to the superior colliculus was increased with relaxing acupressure and decreased with stimulating acupressure, whereas DMN connectivity to the bilateral pulvinar was increased with stimulating and decreased with relaxing acupressure (p < 0.05 corrected). These data suggest that self-administered acupressure at different acupoints has specificity in relation to their mechanisms of action in fatigued breast cancer survivors.

Structural MRI-based detection of Alzheimer's disease using feature ranking and classification error

BACKGROUND AND OBJECTIVE

This paper presents an automatic computer-aided diagnosis (CAD) system based on feature ranking for detection of Alzheimer's disease (AD) using structural magnetic resonance imaging (sMRI) data.

METHODS

The proposed CAD system is composed of four systematic stages. First, global and local differences in the gray matter (GM) of AD patients compared to the GM of healthy controls (HCs) are analyzed using a voxel-based morphometry technique. The aim is to identify significant local differences in the volume of GM as volumes of interests (VOIs). Second, the voxel intensity values of the VOIs are extracted as raw features. Third, the raw features are ranked using a seven-feature ranking method, namely, statistical dependency (SD), mutual information (MI), information gain (IG), Pearson's correlation coefficient (PCC), t-test score (TS), Fisher's criterion (FC), and the Gini index (GI). The features with higher scores are more discriminative. To determine the number of top features, the estimated classification error based on training set made up of the AD and HC groups is calculated, with the vector size that minimized this error selected as the top discriminative feature. Fourth, the classification is performed using a support vector machine (SVM). In addition, a data fusion approach among feature ranking methods is introduced to improve the classification performance.

RESULTS

The proposed method is evaluated using a data-set from ADNI (130 AD and 130 HC) with 10-fold cross-validation. The classification accuracy of the proposed automatic system for the diagnosis of AD is up to 92.48% using the sMRI data.

CONCLUSIONS

An automatic CAD system for the classification of AD based on feature-ranking method and classification errors is proposed. In this regard, seven-feature ranking methods (i.e., SD, MI, IG, PCC, TS, FC, and GI) are evaluated. The optimal size of top discriminative features is determined by the classification error estimation in the training phase. The experimental results indicate that the performance of the proposed system is comparative to that of state-of-the-art classification models.

Evidence of a Synergistic Effect of Acupoint Combination: A Resting-State Functional Magnetic Resonance Imaging Study

Objective: This study aimed to find evidence of a synergistic effect of acupoint combinations by analyzing different brain regions activated after acupuncture at different acupoint combinations.

Methods: A total of 57 healthy subjects were randomly distributed into three groups: LR3 plus KI3 acupoints, LR3 plus sham acupoint, or LR3 alone. They underwent a magnetic resonance imaging scan before and after acupuncture. The amplitude of low-frequency fluctuation (ALFF) and regional homogeneity (ReHo) values of different brain regions were analyzed to observe changes in brain function.

Results: ALFF and ReHo produced an activated area in the cerebellum posterior lobe after acupuncture at LR3 plus KI3 acupoints versus LR3 alone. ALFF and ReHo revealed altered activity in Brodmann area 10 (BA10), BA18, and brainstem pons after acupuncture at LR3 plus sham acupoint compared with at LR3 alone. A comparison of acupuncture at LR3 plus KI3 acupoints with LR3 plus sham acupoint demonstrated an increase in BA6 of ALFF and a downregulation of ReHo.

Conclusions: The increased number of brain regions with altered brain activity after acupuncture at acupoint combinations versus a single acupoint are evidence of the synergistic effect of acupoint combinations. BA6 was significantly activated after acupuncture at LR3 plus KI3 acupoints compared with at LR3 plus sham acupoint, suggesting that BA6 is the specific region of synergistic effect of acupoint combinations of LR3 plus KI3 acupoints. Affected brain regions were different between acupuncture at LR3 plus sham acupoint and LR3 alone, which indicates that the sham acupoint may have some psychological effect. However, the specific mechanism of acupoint combinations requires further research.

The Status of the Quality Control in Acupuncture-Neuroimaging Studies

Using neuroimaging techniques to explore the central mechanism of acupuncture gains increasing attention, but the quality control of acupuncture-neuroimaging study remains to be improved. We searched the PubMed Database during 1995 to 2014. The original English articles with neuroimaging scan performed on human beings were included. The data involved quality control including the author, sample size, characteristics of the participant, neuroimaging technology, and acupuncture intervention were extracted and analyzed. The rigorous inclusion and exclusion criteria are important guaranty for the participants' homogeneity. The standard operation process of acupuncture and the stricter requirement for acupuncturist play significant role in quality control. More attention should be paid to the quality control in future studies to improve the reproducibility and reliability of the acupuncture-neuroimaging studies.

Neural mechanisms of acupuncture as revealed by fMRI studies

As an ancient therapeutic method, acupuncture has been used to treat many diseases as an adjunctive therapy. However, its clinical efficacy remains controversial and the neural mechanisms have not been well understood. Accumulating studies have revealed that fMRI has made it possible to study brain responses to acupuncture. This review aims to provide scientific evidence to support the notion and discuss how these findings contribute to the neural mechanisms of acupuncture.

Acupuncture at the Taixi (KI3) acupoint activates cerebral neurons in elderly patients with mild cognitive impairment

Our previous findings have demonstrated that acupuncture at the Taixi (KI3) acupoint in healthy youths can activate neurons in cognitive-related cerebral cortex. Here, we investigated whether acupuncture at this acupoint in elderly patients with mild cognitive impairment can also activate neurons in these regions. Resting state and task-related functional magnetic resonance imaging showed that the pinprick senstation of acupuncture at the Taixi acupoint differed significantly between elderly patients with mild cognitive impairment and healthy elderly controls. Results showed that 20 brain regions were activated in both groups of participants, including the bilateral anterior cingulate gyrus (Brodmann areas [BA] 32, 24), left medial frontal cortex (BA 9, 10, 11), left cuneus (BA 19), left middle frontal gyrus (BA 11), left lingual gyrus (BA 18), right medial frontal gyrus (BA 11), bilateral inferior frontal gyrus (BA 47), left superior frontal gyrus (BA11), right cuneus (BA 19, 18), right superior temporal gyrus (BA 38), left subcallosal gyrus (BA 47), bilateral precuneus (BA 19), right medial frontal gyrus (BA 10), right superior frontal (BA 11), left cingulate gyrus (BA 32), left precentral gyrus (BA 6), and right fusiform gyrus (BA 19). These results suggest that acupuncture at the Taixi acupoint in elderly patients with mild cognitive impairment can also activate some brain regions.

Multivariate granger causality analysis of acupuncture effects in mild cognitive impairment patients: an FMRI study

Evidence from clinical reports has indicated that acupuncture has a promising effect on mild cognitive impairment (MCI). However, it is still unknown that by what way acupuncture can modulate brain networks involving the MCI. In the current study, multivariate Granger causality analysis (mGCA) was adopted to compare the interregional effective connectivity of brain networks by varying needling depths (deep acupuncture, DA; superficial acupuncture, SA) and at different cognitive states, which were the MCI and healthy control (HC). Results from DA at KI3 in MCI showed that the dorsolateral prefrontal cortex and hippocampus emerged as central hubs and had significant causal influences with each other, but significant in HC for DA. Moreover, only several brain regions had remarkable causal interactions following SA in MCI and even few brain regions following SA in HC. Our results indicated that acupuncture at KI3 at different cognitive states and with varying needling depths may induce distinct reorganizations of effective connectivities of brain networks, and DA at KI3 in MCI can induce the strongest and more extensive effective connectivities related to the therapeutic effect of acupuncture in MCI. The study demonstrated the relatively functional specificity of acupuncture at KI3 in MCI, and needling depths play an important role in acupuncture treatments.