Tuesday, November 29, 2022

Classification Of Parkinson’s Disease

Sidebar: Morris K Udall Centers Of Excellence For Parkinson’s Disease Research

Are there different types of Parkinson’s?

The Morris K. Udall Parkinsons Disease Research Act of 1997 authorized the NIH to greatly accelerate and expand PD research efforts by launching the NINDS Udall Centers of Excellence, a network of research centers that provide a collaborative, interdisciplinary framework for PD research. Udall Center investigators, along with many other researchers funded by the NIH, have made substantial progress in understanding PD, including identifying disease-associated genes investigating the neurobiological mechanisms that contribute to PD, developing and improving PD research models, and discovering and testing potential therapeutic targets for developing novel treatment strategies.

The Udall Centers continue to conduct critical basic, translational, and clinical research on PD including: 1) identifying and characterizing candidate and disease-associated genes, 2) examining neurobiological mechanisms underlying the disease, and 3) developing and testing potential therapies. As part of the program, Udall Center investigators work with local communities of patients and caregivers to identify the challenges of living with PD and to translate scientific discoveries into patient care. The Centers also train the next generation of physicians and scientists who will advance our knowledge of and treatments for PD. See the full list of Udall Centers.

Data Types And Associated Outcomes

Out of 209 studies, 122 applied machine learning methods to movement-related data, i.e., voice recordings , movement data , or handwritten patterns . Imaging modalities analyzed including MRI , SPECT , and positron emission tomography . Five studies analyzed CSF samples . In 18 studies , a combination of different types of data was used.

Ten studies used data that do not belong to any categories mentioned above, such as single nucleotide polymorphisms , electromyography , OCT , cardiac scintigraphy , Patient Questionnaire of Movement Disorder Society Unified Parkinson’s Disease Rating Scale , whole-blood gene expression profiles , transcranial sonography , eye movements , electroencephalography , and serum samples .

Given that studies used different data modalities and sources, and sometimes different samples of the same database, a summary of model performance, instead of direct comparison across studies, is provided.

Voice Recordings

Data Source And Sample Size

In 93 out of 209 studies , original data were collected from human participants. In 108 studies , data used were from public repositories and databases, including University of California at Irvine Machine Learning Repository , Parkinson’s Progression Markers Initiative , PhysioNet , HandPD dataset , mPower database , and 6 other databases .

You May Like: Prayer For Parkinson’s Disease

What Are The Different Forms Of Parkinsonism

There are three main forms of parkinsonism, as well as other related conditions.

Most people with parkinsonism have idiopathic Parkinsons disease, also known as Parkinsons. Idiopathic means the cause is unknown.

The most common symptoms of idiopathic Parkinsons are tremor, rigidity and slowness of movement.

Vascular parkinsonism affects people with restricted blood supply to the brain. Sometimes people who have had a mild stroke may develop this form of parkinsonism.

Common symptoms include problems with memory, sleep, mood and movement.

Some drugs can cause parkinsonism.

Neuroleptic drugs , which block the action of the chemical dopamine in the brain, are thought to be the biggest cause of drug-induced parkinsonism.

The symptoms of drug-induced parkinsonism tend to stay the same only in rare cases do they progress in the way that Parkinsons symptoms do.

Drug-induced parkinsonism only affects a small number of people, and most will recover within months and often within days or weeks of stopping the drug thats causing it.

Feature Analysis And Selection

Distribution of features of parkinson

Feature selection was the next step to assess the potential of the extracted metrics and to select the most suitable parameters for machine learning classifiers to obtain the highest accuracy. From a statistical point of view, the Spearmans correlation was estimated with SPSS21 to determine the correlation between measured biomechanical parameters and clinical scores estimated by the neurologist. As proposed in , features with a correlation higher than strong were considered as significant with respect to the clinical scores. Since the measurement for the PwPD group violated the normality assumption, the unpaired MannWhitney U test for non-parametric samples was calculated to identify the ideal features for comparison between healthy controls and the PD group. The most dominant side of PD participants was compared with the participants best performing side of the healthy control group. The choice of the cut-off points to accept the alternative hypothesis is totally subjective. The common use of p0.05 was chosen by Fisher . A probability value of p< 0.05 was considered significant for all the analyses . ANOVA tests were used to assess differences in objective measurements between PwPD and control groups. Obtained results from ANOVA were presented in mean and standard deviation. All the statistical analyses were performed with SPSS21 . Simultaneously, Cohens d effect size was also calculated with an available online calculator .

Read Also: Does Vitamin B12 Help Parkinson’s

The Relation Between Levodopa Treatment And Pain In Parkinson’s Disease

Levodopa is the most widely used and effective medication in the treatment of motor signs of PD. However, when used over a number of years, virtually all typical patients will present with complications related to pharmacokinetic and pharmacodynamic characteristics. The best-known of these include motor fluctuations such as WO, in which the therapeutic effects of levodopa do not last as long as they previously did, and dyskinesias . Although only recently recognized, nonmotor fluctuations occur in a similar fashion in nearly all patients and can be main determinants of disability and worse quality of life . NMS can be divided into three categories: neuropsychiatric, autonomic, and sensory. While most of these manifest during off periods and WO, some, such as agitation, psychosis, diaphoresis, and pain, can manifest when levodopa levels are at a peak. In this situation, pain appears to be more related to involuntary movements than to dopamine levels per se. Fluctuation of sensory symptoms can be due to failure of primary somatosensory mechanisms .

Flowchart Of The Proposed Algorithm

For simplification, the proposed algorithm was called PD_MEdit_EL and a flowchart can be seen below . The PD_MEdit_EL algorithm includes two major parts, with one part optimizing speech samples via MENN while the other performs the classification using the ensemble learning algorithm, with either RF or DNNE used. Chronologically, training speech samples are optimized using the MENN algorithm, these training samples are then processed with the ensemble learning algorithm and a trained learning model is obtained. Finally, the trained learning algorithm is applied to aid in test sample classification.

Fig. 1

You May Like: Similar To Parkinsons

Assessment And Classification Of Pain In Pd

The prevalence of pain in PD can vary from 34% to 83% depending on methodological assessments. Possible explanations for this large range include the tools and criteria used for the diagnosis and profile of the population studied. For instance, while some studies only considered this symptom to be present when it lasted more than three months , others did not specify any time criterion . Among the diagnostic tools for the assessment of pain in PD, the most widely used is the Brief Pain Inventory however, because it only analyzes pain in the previous 24 hours, it can underestimate its prevalence. As pain is an unpleasant sensory and emotional experience , the use of a multidimensional scale to evaluate it is recommended . The widely used McGill Pain Questionnaire is an example of this type of scale, evaluating sensory-discriminatory and affective-motivational domains .

Nonetheless, there is a lack of consensus regarding the assessment and classification of pain in PD patients. The first classification, published by Ford , is the most commonly used, but other proposals have been put forward as a result of the growing interest in the subject .

Level Of Significance Of Pd: Medit: El Algorithm

Analysis on different types of Parkinson’s Disease

In an attempt to establish that a significant difference is present between the PD_MEdit_EL algorithm and the other examined algorithms, the p values of the ACCs, SENs and SPEs between the algorithms based on ten times experiments were calculated . In terms of LOO and LOSO, the differences between the RF and DNNE when compared to the SVM or SVM showed some highly significant differences. While these results show a significant difference between the PD_MEdit_EL algorithm and those examined, a significant difference was not noted when comparing the RF and DNNE , thus suggesting that the two algorithms would perform in a comparable fashion.

Table 7 Improvement with the MENN algorithm for the Training_Data set

Statistically significant differences were also examined using the Test_Data set . The same trend was seen with this dataset, with some highly significant differences noted between the RF and DNNE when compared to the SVM or SVM . However, when comparing the RF and DNNE , no significant difference was noted for subjects, but the two were significantly statistically different for samples.

Table 8 Improvement with the MENN algorithm for the Test_Data set

The main contributions and innovations of this paper can be described as follows:

  • 1.
  • Recommended Reading: Diseases Similar To Parkinsons

    Types Of Parkinsons Disease

    Parkinsons disease is caused by the dysfunction and death of dopamine-producing neurons in the brain. Parkinsonism is a broader term that encompasses Parkinsons disease itself, as well as other conditions that cause motor symptoms similar to those seen in Parkinsons, like tremor and abnormally slow movement.

    Sidebar: Ninds Steps Up Pursuit Of Pd Biomarkers

    In 2012, the NINDS dramatically accelerated efforts to identify biomarkers by establishing the Parkinsons Disease Biomarkers Program . This unprecedented program unites a range of stakeholders from basic and clinical researchers to healthcare professionals, the NINDS staff, information technology experts, and people living with PD and their families.

    PDBP supports research and builds resources aimed at accelerating the discovery of biomarkers to ultimately slow the progression of PD. For example, the program has established a repository of biological specimens and a Data Management Resource system maintained by the NIH Center for Information Technology. The DMR allows researchers to access clinical, imaging, genetic, and biologic data, while a complementary PDBP-supported project develops statistical tools to analyze vast quantities of data so that patterns can be identified across these diverse sources of information.

    You May Like: On And Off Phenomenon

    Classification Algorithms For Healthy Control And Pwpd

    Supervised learning methods such as support vector machine , logistic regression , and naive Bayes with tenfold cross validation were used to classify both groups of subjects . The tenfold cross validation randomly splits the n different subjects into tenfolds with roughly proportional numbers of healthy and PwPD in each fold. The prediction algorithm is repeated 10 times with the cases of each fold withheld from the training set in turn, the cross-validated error rate being the average error rate on the withheld cases. A typical fold contained 10 subjects for the prostate data, healthy and PwPD, which were then predicted by the rule constructed from the data of the other remaining subjects. The SVM classifier was trained with sequential minimal optimization methods and with polynomial kernel. All the classifiers were developed under the environment of machine learning software weka3.6 .

    Title: Parkinsonian Chinese Speech Analysis Towards Automatic Classification Of Parkinson’s Disease

    Parkinsons disease

    Abstract: Speech disorders often occur at the early stage of Parkinson’s disease .The speech impairments could be indicators of the disorder for early diagnosis,while motor symptoms are not obvious. In this study, we constructed a newspeech corpus of Mandarin Chinese and addressed classification of patients withPD. We implemented classical machine learning methods with ranking algorithmsfor feature selection, convolutional and recurrent deep networks, and an end toend system. Our classification accuracy significantly surpassedstate-of-the-art studies. The result suggests that free talk has strongerclassification power than standard speech tasks, which could help the design offuture speech tasks for efficient early diagnosis of the disease. Based onexisting classification methods and our natural speech study, the automaticdetection of PD from daily conversation could be accessible to the majority ofthe clinical population.

    Read Also: Fitflop Shoes For Parkinson’s

    Classification Across All Genotypes

    Next, we asked whether the classifier was able to assign individual flies to the correct genotype after training on the entire 10-class dataset. Performance is expected to drop on this type of task but the chance classification rate is also lower .

    shows the result of performing this analysis using either unregularized or regularized data . The data are plotted as box plots with the notches indicating the interquartile ranges. Surprisingly, even the unregularized datasets achieved a classification performance that was significantly above chance . The PINK15, DJ-172, w1118 and wDah genotypes were most distinct with the classifier performing close to chance on the w1 genotypes.

    Figure 6

    Spatiotemporal response profiles allow accurate genotyping of individual flies.

    Following discriminant analysis machine learning of the entire 10-genotype dataset , individual flies were accurately classified into the correct genotypic class using the raw, un-processed data. All flies were classified correctly at a level significantly greater than chance . Regularizing the data significantly improved classification accuracies for all genotypes. All classifications were now greater than chance at p< .001 with the best performances approaching 60% accuracy.

    Drosophila Stocks And Maintenance

    Responses were measured from ten different Drosophila genotypes. We chose five PD genotypesfour with early-onset PD-related mutations and one with a loss-of-function mutation in the Drosophila LRRK gene associated with late-onset PD. All the mutant strains had white eyes. Our control group consisted of four different wildtype white eyed strains originating in different laboratories. As an additional control we tested a well characterised model of non-PD neurodegeneration, a mutation in the eggroll gene . eggroll1, w1118, w1, PINK15 and dLrrkEx1 fly stocks were obtained from the Bloomington Drosophila Stock Center . PINK1B9, DJ-172 and DJ-193 stocks were generously gifted by Dr Alex Whitworth . wDahomey flies were a kind gift from Dr Susan Broughton . The second w1118 stock was a kind gift from Dr Tobias Rasse . The DJ-172,DJ-193, dLrrkEx1and eggroll1were tested as homozygotes. The PINK15 and PINK1B9 were tested as hemizygotes as the gene is on the X-chromosome. All D. melanogaster lines were raised in a 12hr:12hr light:dark cycle at 25°C on standard cornmeal-yeast-sucrose medium.

    Don’t Miss: Does Sam Waterston Have Parkinsons

    Parkinson’s Disease And Parkinsonism

    Parkinson’s disease is named for Dr. James Parkinson, who in 1817 first described the features of this illness. Features of Parkinson’s disease include tremor, slow movement , and rigid muscles . People with parkinsonism may have Parkinson’s disease or another illness with similar symptoms.

    Other conditions and diseases that cause parkinsonism may also cause symptoms that are not seen with Parkinson’s disease. These conditions may be treated differently than Parkinson’s disease. Unlike Parkinson’s, some conditions that cause parkinsonism are reversible.

    • Parkinson’s-plus syndromes are a group of disorders characterized by the degeneration of nerve cells in different parts of the brain. They include progressive supranuclear palsy , corticobasal degeneration , and multiple system atrophy , among others. Parkinson’s-plus syndromes have parkinsonian features as well as features that are not associated with Parkinson’s disease. These syndromes usually respond poorly to levodopa or dopamine agonists.
    • Secondary or symptomatic parkinsonism describes the syndrome of parkinsonism when it occurs as the result of an identifiable cause. For example, certain medicines, brain tumors, strokes, infections , and toxins can cause secondary parkinsonism.

    How Is Parkinsonism Diagnosed

    What are the different forms and stages of Parkinson’s disease?

    You should be referred to a Parkinsons specialist for the diagnosis of any parkinsonism. They may wish to explore different things before giving you a diagnosis.

    Your specialist will look at your medical history, ask you about your symptoms and do a medical examination.

    Telling the difference between types of parkinsonism isnt always easy, for the following reasons:

    • The first symptoms of the different forms of parkinsonism are so similar.
    • In many cases, parkinsonism develops gradually. Symptoms that allow your doctor to make a specific diagnosis may only appear as your condition progresses.
    • Everyone with parkinsonism is different and has different symptoms.

    Find out more: see our information on symptoms of Parkinsons, and diagnosing Parkinsons.

    One of the most useful tests to find out what sort of parkinsonism you may have is to see how you respond to treatment.

    If your specialist thinks you have idiopathic Parkinsons, theyll expect you to have a good response to Parkinsons drugs such as levodopa . A good response means that your symptoms will improve. Sometimes, it will only be clear that youve responded to medication when the drug is reduced or stopped, and your symptoms become more obvious again.

    If you dont have any response to Parkinsons medication, your specialist will have to look again at your diagnosis.

    Although not routinely available, your specialist may wish to carry out some of the tests below.

    Current tests available include:

    Don’t Miss: Judy Woodruff Health Problems

    Tunable Q Wavelet Transform Based Emotion Classification In Parkinsons Disease Using Electroencephalography

    • Roles Funding acquisition, Investigation, Project administration

      ¶These authors also contributed equally to this work.

      Affiliation Department of Physiology, Faculty of Medicine, Kuwait University, Kuwait City, Kuwait

    • Roles Funding acquisition, Investigation, Project administration

      ¶These authors also contributed equally to this work.

      Affiliation Department of Physiology, Faculty of Medicine, Kuwait University, Kuwait City, Kuwait

    • Contributed equally to this work with: Smith K. Khare, Varun Bajaj

      Roles Visualization, Writing original draft, Writing review & editing

      Affiliation Department of Electronics and Communication, PDPM Indian Institute of Information Technology, Design and Manufacturing, Jabalpur, India

    Associations Between Clinical And 360 Turning Characteristics Of People With Pd

    Furthermore, clinical characteristics related to PD severity, such as UPDRS total and III scores, PIGD score, Hoehn and Yahr stage, and NFOGQ score, were identified as the indicators of FOG . Previous studies have shown the association of the severity of FOG with motor deficit . It has been suggested that induced motor deficit such as the loss of automaticity along with stepping inhibition during turning led to repeated weight shifts without stepping, resulting in trembling of limbs related to FOG . In particular, our result indicated that the outer step length decreased as the NFOGQ score increased in freezers. In this study, no difference between the inner and outer step lengths in freezers was observed during the turning task. These results do not indicate the asymmetry of steps during turning in freezers with advanced disease severity , which may be reflected as reduced normal asymmetric gait strategy and bilateral motor coordination during turning .

    You May Like: Sam Waterston Parkinson’s

    Popular Articles
    Related news