The use of a color coded probability scale to interpret smell tests in suspected parkinsonism

Long Short-Term Memory-Based Deep Learning Models for Screening Parkinson's Disease Using Sequential Diagnostic Codes

BACKGROUND AND PURPOSE

It is challenging to detect Parkinson's disease (PD) in its early stages, which has prompted researchers to develop techniques based on machine learning methods for detecting PD. However, previous studies did not fully incorporate the slow progression of PD over a long period of time nor consider that its symptoms occur in a time-sequential manner. Contributing to the literature on PD, which has relied heavily on cross-sectional data, this study aimed to develop a method for detecting PD early that can process time-series information using the long short-term memory (LSTM) algorithm.

METHODS

We sampled 926 patients with PD and 9,260 subjects without PD using medical-claims data. The LSTM algorithm was tested using diagnostic histories, which contained the diagnostic codes and their respective time information. We compared the prediction power of the 12-month diagnostic codes under two different settings over the 4 years prior to the first PD diagnosis.

RESULTS

The model that was trained using the most-recent 12-month diagnostic codes had the best performance, with an accuracy of 94.25%, a sensitivity of 82.91%, and a specificity of 95.26%. The other three models (12-month codes from 2, 3, and 4 years prior) were found to have comparable performances, with accuracies of 92.27%, 91.86%, and 91.81%, respectively. The areas under the curve from our data settings ranged from 0.839 to 0.923.

CONCLUSIONS

We explored the possibility that PD specialists could benefit from our proposed machine learning method as an early detection method for PD.

An Unsupervised Neural Network Feature Selection and 1D Convolution Neural Network Classification for Screening of Parkinsonism

Parkinson’s disease (PD) is the second most common neurodegenerative disorder after Alzheimer’s disease. It has a slow progressing neurodegenerative disorder rate. PD patients have multiple motor and non-motor symptoms, including vocal impairment, which is one of the main symptoms. The identification of PD based on vocal disorders is at the forefront of research. In this paper, an experimental study is performed on an open source Kaggle PD speech dataset and novel comparative techniques were employed to identify PD. We proposed an unsupervised autoencoder feature selection technique, and passed the compressed features to supervised machine-learning (ML) algorithms. We also investigated the state-of-the-art deep learning 1D convolutional neural network (CNN-1D) for PD classification. In this study, the proposed algorithms are support vector machine, logistic regression, random forest, naïve Bayes, and CNN-1D. The classifier performance is evaluated in terms of accuracy score, precision, recall, and F1 score measure. The proposed 1D-CNN model shows the highest result of 0.927%, and logistic regression shows 0.922% on the benchmark dataset in terms of F1 measure. The major contribution of the proposed approach is that unsupervised neural network feature selection has not previously been investigated in Parkinson’s detection. Clinicians can use these techniques to analyze the symptoms presented by patients and, based on the results of the above algorithms, can diagnose the disease at an early stage, which will allow for improved future treatment and care.

Combined assessment by transcranial sonography and Sniffin' Sticks test has a similar diagnostic accuracy compared to brain SPECT for Parkinson's disease diagnosis

OBJECTIVES

This study aimed to investigate the accuracy of TCS combined with the Sniffin' sticks olfactory test (SST-16) for differentiation between idiopathic PD patients and healthy controls compared to that of ^99 mTc-TRODAT-1 SPECT (TRODAT).

METHODS

A cross-sectional study included PD patients diagnosed in accordance with United Kingdom PD Society Brain Bank criteria and a control group of age and sex- matched healthy subjects. All patients were examined by a movement disorder specialist and underwent brain SPECT using TRODAT, TCS examination and SST-16 test. Receiver Operating Characteristic (ROC) curves were used to calculate cut-off points for TCS, striatal TRODAT binding potentials and SST-16. The area under the ROC curve determined the diagnostic accuracy of the method.

RESULTS

Twenty patients with PD (13 males and 7 females) and nine healthy subjects were included. Median age of PD onset was 56.5 years with median disease duration of 5 years. A larger substantia nigra (SN) echogenic area was observed in the PD group (p = 0.013). SN echogenic area cut-off point of 0.22 cm² was obtained from a ROC curve for PD diagnosis. Considering this cut-off point, TCS diagnostic accuracy was estimated at 79.2% for PD diagnosis. The cut-off value of 0.90 for striatal TRODAT binding was associated with 99% diagnostic accuracy for the diagnosis of PD. SST-16 values equal or less than 9 points showed an 85.8% diagnostic accuracy for PD diagnosis. Combination of both SST-16 and TCS improved the diagnostic accuracy to 95% for PD diagnosis.

CONCLUSION

Combined SST-16 and TCS assessment was indicated as accurate for distinguishing PD patients from healthy controls. The diagnostic accuracy of TCS combined with SST-16 for differentiation between idiopathic PD patients and healthy controls is similar to that of SPECT TRODAT.

Smell tests can discriminate Parkinson's disease patients from healthy individuals: A meta-analysis

BACKGROUND

Olfactory impairment is common in Parkinson's disease (PD). The authors aimed to identify the clinical tests used to assess olfactory function and examine their ability to distinguish PD with different disease duration from healthy individuals with physiological aging.

METHODS

Cross-sectional studies published until May 2020 that assessed the olfaction of individuals with PD using search terms related to PD, olfactory function, and assessment were searched on PubMed, PsycInfo, Cinahl, and Web of Science databases.

RESULTS

Twelve smell tests were identified from the reviewed studies (n = 125) that assessed 8776 individuals with PD. Data of 6593 individuals with PD and 8731 healthy individuals were included in the meta-analyses. Individuals with PD presented worse performance than healthy individuals, regardless of the smell test used. The University of Pennsylvania Smell Identification Test (UPSIT) was used by most studies (n = 2310 individuals with PD) and presented smaller heterogeneity. When the studies were subclassified according to the years of PD duration, there were no significant differences.

CONCLUSION

All smell tests were able to discriminate the olfactory function of PD from that of healthy individuals, although the UPSIT was widely used. The abnormal olfaction was not related to the disease duration. Systematic review protocol registration (PROSPERO/2020-CRD42020160878).

Comparison between four published definitions of hyposmia in Parkinson's disease

OBJECTIVES

Hyposmia is a common feature of Parkinson's disease (PD), yet there is no standard method to define it. A comparison of four published methods was performed to explore and highlight differences.

MATERIALS AND METHODS

Olfactory testing was performed in 2097 cases of early PD in two prospective studies. Olfaction was assessed using various cut-offs, usually corrected by age and/or gender. Control data were simulated based on the age and gender structure of the PD cases and published normal ranges. Association with age, gender, and disease duration was explored by method and study cohort. Prevalence of hyposmia was compared with the age and gender-matched simulated controls. Between method agreement was measured using Cohen's kappa and Gwet's AC1.

RESULTS

Hyposmia was present in between 69.1% and 97.9% of cases in Tracking Parkinson's cases, and between 62.2% and 90.8% of cases in the Parkinson's Progression Marker Initiative, depending on the method. Between-method agreement varied (kappa 0.09-0.80, AC1 0.55-0.86). The absolute difference between PD cases and simulated controls was similar for men and women across methods. Age and male gender were positively associated with hyposmia (p < .001, all methods). Odds of having hyposmia increased with advancing age (OR:1.06, 95% CI:1.03, 1.10, p < .001). Longer disease duration had a negative impact on overall olfactory performance.

CONCLUSIONS

Different definitions of hyposmia give different results using the same dataset. A standardized definition of hyposmia in PD is required, adjusting for age and gender, to account for the background decline in olfactory performance with ageing, especially in men.

Early detection of Parkinson's disease through patient questionnaire and predictive modelling

Early detection of Parkinson's disease (PD) is important which can enable early initiation of therapeutic interventions and management strategies. However, methods for early detection still remain an unmet clinical need in PD. In this study, we use the Patient Questionnaire (PQ) portion from the widely used Movement Disorder Society-Unified Parkinson's Disease Rating Scale (MDS-UPDRS) to develop prediction models that can classify early PD from healthy normal using machine learning techniques that are becoming popular in biomedicine: logistic regression, random forests, boosted trees and support vector machine (SVM). We carried out both subject-wise and record-wise validation for evaluating the machine learning techniques. We observe that these techniques perform with high accuracy and high area under the ROC curve (both >95%) in classifying early PD from healthy normal. The logistic model demonstrated statistically significant fit to the data indicating its usefulness as a predictive model. It is inferred that these prediction models have the potential to aid clinicians in the diagnostic process by joining the items of a questionnaire through machine learning.

Multi-modal Brain MRI in Subjects with PD and iRBD

Idiopathic rapid eye movement sleep behavior disorder (iRBD) is a condition that often evolves into Parkinson's disease (PD). Therefore, by monitoring iRBD it is possible to track the neurodegeneration of individuals who may progress to PD. Here we aimed at piloting the characterization of brain tissue properties in mid-brain subcortical regions of 10 healthy subjects, 8 iRBD, and 9 early-diagnosed PD. We used a battery of magnetic resonance imaging (MRI) contrasts at 3 T, including adiabatic and non-adiabatic rotating frame techniques developed by our group, along with diffusion tensor imaging (DTI) and resting-state fMRI. Adiabatic T_1ρ and T_2ρ, and non-adiabatic RAFF4 (Relaxation Along a Fictitious Field in the rotating frame of rank 4) were found to have lower coefficient of variations and higher sensitivity to detect group differences as compared to DTI parameters such as fractional anisotropy and mean diffusivity. Significantly longer T_1ρ were observed in the amygdala of PD subjects vs. controls, along with a trend of lower functional connectivity as measured by regional homogeneity, thereby supporting the notion that amygdalar dysfunction occurs in PD. Significant abnormalities in reward networks occurred in iRBD subjects, who manifested lower network strength of the accumbens. In agreement with previous studies, significantly longer T_1ρ occurred in the substantia nigra compacta of PD vs. controls, indicative of neuronal degeneration, while regional homogeneity was lower in the substantia nigra reticulata. Finally, other trend-level findings were observed, i.e., lower RAFF4 and T_2ρ in the midbrain of iRBD subjects vs. controls, possibly indicating changes in non-motor features as opposed to motor function in the iRBD group. We conclude that rotating frame relaxation methods along with functional connectivity measures are valuable to characterize iRBD and PD subjects, and with proper validation in larger cohorts may provide pathological signatures of iRBD and PD.

Equating scores of the University of Pennsylvania Smell Identification Test and Sniffin' Sticks test in patients with Parkinson's disease

BACKGROUND

Impaired olfaction is an important feature in Parkinson's disease (PD) and other neurological diseases. A variety of smell identification tests exist such as "Sniffin' Sticks" and the University of Pennsylvania Smell Identification Test (UPSIT). An important part of research is being able to replicate findings or combining studies in a meta-analysis. This is difficult if olfaction has been measured using different metrics. We present conversion methods between the: UPSIT, Sniffin' 16, and Brief-SIT (B-SIT); and Sniffin' 12 and Sniffin' 16 odour identification tests.

METHODS

We used two incident cohorts of patients with PD who were tested with either the Sniffin' 16 (n = 1131) or UPSIT (n = 980) and a validation dataset of 128 individuals who took both tests. We used the equipercentile and Item Response Theory (IRT) methods to equate the olfaction scales.

RESULTS

The equipercentile conversion suggested some bias between UPSIT and Sniffin' 16 tests across the two groups. The IRT method shows very good characteristics between the true and converted Sniffin' 16 (delta mean = 0.14, median = 0) based on UPSIT. The equipercentile conversion between the Sniffin' 12 and 16 item worked well (delta mean = 0.01, median = 0). The UPSIT to B-SIT conversion showed evidence of bias but amongst PD cases worked well (mean delta = -0.08, median = 0).

CONCLUSION

We have demonstrated that one can convert UPSIT to B-SIT or Sniffin' 16, and Sniffin' 12 to 16 scores in a valid way. This can facilitate direct comparison between tests aiding future collaborative analyses and evidence synthesis.

Assessment of Olfactory Function in MAPT-Associated Neurodegenerative Disease Reveals Odor-Identification Irreproducibility as a Non-Disease-Specific, General Characteristic of Olfactory Dysfunction

Olfactory dysfunction is associated with normal aging, multiple neurodegenerative disorders, including Parkinson's disease, Lewy body disease and Alzheimer's disease, and other diseases such as diabetes, sleep apnea and the autoimmune disease myasthenia gravis. The wide spectrum of neurodegenerative disorders associated with olfactory dysfunction suggests different, potentially overlapping, underlying pathophysiologies. Studying olfactory dysfunction in presymptomatic carriers of mutations known to cause familial parkinsonism provides unique opportunities to understand the role of genetic factors, delineate the salient characteristics of the onset of olfactory dysfunction, and understand when it starts relative to motor and cognitive symptoms. We evaluated olfactory dysfunction in 28 carriers of two MAPT mutations (p.N279K, p.P301L), which cause frontotemporal dementia with parkinsonism, using the University of Pennsylvania Smell Identification Test. Olfactory dysfunction in carriers does not appear to be allele specific, but is strongly age-dependent and precedes symptomatic onset. Severe olfactory dysfunction, however, is not a fully penetrant trait at the time of symptom onset. Principal component analysis revealed that olfactory dysfunction is not odor-class specific, even though individual odor responses cluster kindred members according to genetic and disease status. Strikingly, carriers with incipient olfactory dysfunction show poor inter-test consistency among the sets of odors identified incorrectly in successive replicate tests, even before severe olfactory dysfunction appears. Furthermore, when 78 individuals without neurodegenerative disease and 14 individuals with sporadic Parkinson's disease were evaluated twice at a one-year interval using the Brief Smell Identification Test, the majority also showed inconsistency in the sets of odors they identified incorrectly, independent of age and cognitive status. While these findings may reflect the limitations of these tests used and the sample sizes, olfactory dysfunction appears to be associated with the inability to identify odors reliably and consistently, not with the loss of an ability to identify specific odors. Irreproducibility in odor identification appears to be a non-disease-specific, general feature of olfactory dysfunction that is accelerated or accentuated in neurodegenerative disease. It may reflect a fundamental organizational principle of the olfactory system, which is more "error-prone" than other sensory systems.

Tracking Parkinson's: Study Design and Baseline Patient Data

Background:

There is wide variation in the phenotypic expression of Parkinson’s disease (PD), which is driven by both genetic and epidemiological influences.

Objectives:

To define and explain variation in the clinical phenotype of PD, in relation to genotypic variation.

Methods:

Tracking Parkinson’s is a multicentre prospective longitudinal epidemiologic and biomarker study of PD. Patients attending specialist clinics in the United Kingdom with recent onset (<3.5 years) and young onset (diagnosed <50 years of age) PD were enrolled. Motor, non-motor and quality of life assessments were performed using validated scales. Cases are followed up 6 monthly up to 4.5 years for recent onset PD, and up to 1 year for young onset PD. We present here baseline clinical data from this large and demographically representative cohort.

Results:

2247 PD cases were recruited (1987 recent onset, 260 young onset). Recent onset cases had a mean (standard deviation, SD) age of 67.6 years (9.3) at study entry, 65.7% males, with disease duration 1.3 years (0.9), MDS-UPDRS 3 scores 22.9 (12.3), LEDD 295 mg/day (211) and PDQ-8 score 5.9 (4.8). Young onset cases were 53.5 years old (7.8) at study entry, 66.9% male, with disease duration 10.2 years (6.7), MDS-UPDRS 3 scores 27.4 (15.3), LEDD 926 mg/day (567) and PDQ-8 score 11.6 (6.1).

Conclusions:

We have established a large clinical PD cohort, consisting of young onset and recent onset cases, which is designed to evaluate variation in clinical expression, in relation to genetic influences, and which offers a platform for future imaging and biomarker research.

High-Accuracy Detection of Early Parkinson's Disease through Multimodal Features and Machine Learning

Early (or preclinical) diagnosis of Parkinson's disease (PD) is crucial for its early management as by the time manifestation of clinical symptoms occur, more than 60% of the dopaminergic neurons have already been lost. It is now established that there exists a premotor stage, before the start of these classic motor symptoms, characterized by a constellation of clinical features, mostly non-motor in nature such as Rapid Eye Movement (REM) sleep Behaviour Disorder (RBD) and olfactory loss. In this paper, we use the non-motor features of RBD and olfactory loss, along with other significant biomarkers such as Cerebrospinal fluid (CSF) measurements and dopaminergic imaging markers from 183 healthy normal and 401 early PD subjects, as obtained from the Parkinson's Progression Markers Initiative (PPMI) database, to classify early PD subjects from normal using Naïve Bayes, Support Vector Machine (SVM), Boosted Trees and Random Forests classifiers. We observe that SVM classifier gave the best performance (96.40% accuracy, 97.03% sensitivity, 95.01% specificity, and 98.88% area under ROC). We infer from the study that a combination of non-motor, CSF and imaging markers may aid in the preclinical diagnosis of PD.

Parkinson's disease detection using olfactory loss and REM sleep disorder features

In Parkinson's disease, there exists a prodromal or a premotor phase characterized by symptoms like olfactory loss and sleep disorders, which may last for years or even decades before the onset of motor clinical symptoms. Diagnostic tools based on machine learning using these features can be very useful as they have the potential in early diagnosis of the disease. In the paper, we use olfactory loss feature from 40-item University of Pennsylvania Smell Identification Test (UPSIT) and Sleep behavior disorder feature from Rapid eye movement sleep Behavior Disorder Screening Questionnaire (RBDSQ), obtained from the Parkinson's Progression Marker's Initiative (PPMI) database, to develop automated diagnostic models using Support Vector Machine (SVM) and classification tree methods. The advantage of using UPSIT and RBDSQ is that they are quick, cheap, and can be self-administered. Results show that the models performed with high accuracy and sensitivity, and that they have the potential to aid in early diagnosis of Parkinson's disease.

Clinical prodromes of neurodegeneration in Anderson-Fabry disease

Objective:

To estimate the prevalence of prodromal clinical features of neurodegeneration in patients with Anderson-Fabry disease (AFD) in comparison to age-matched controls.

Methods:

This is a single-center, prospective, cross-sectional study in 167 participants (60 heterozygous females and 50 hemizygous males with genetically confirmed AFD, 57 age-matched controls) using a clinical screening program consisting of structured interview, quantitative tests of motor function, and assessments of cognition, depression, olfaction, orthostatic intolerance, pain, REM sleep behavior disorder, and daytime sleepiness.

Results:

In comparison to age-matched controls (mean age 48.3 years), patients with AFD (mean age 49.0 years) showed slower gait and transfer speed, poorer fine manual dexterity, and lower hand speed, which was independent of focal symptoms due to cerebrovascular disease. Patients with AFD were more severely affected by depression, pain, and daytime sleepiness and had a lower quality of life. These motor and nonmotor manifestations significantly correlated with clinical disease severity. However, patients with AFD did not reveal extrapyramidal motor features or signs of significant cognitive impairment, hyposmia, orthostatic intolerance, or REM sleep behavior disorder, which commonly precede later neurodegenerative disease. In our cohort, there were no differences in neurologic manifestations of AFD between heterozygous females and hemizygous males.

Conclusions:

Aside from cerebrovascular manifestations and small fiber neuropathy, AFD results in a distinct neurologic phenotype comprising poorer motor performance and specific nonmotor features. In contrast to functional loss of glucocerebrosidase in Gaucher disease, α-galactosidase deficiency in AFD is not associated with a typical cluster of clinical features prodromal for neurodegenerative diseases, such as Parkinson disease.

Premotor biomarkers for Parkinson's disease - a promising direction of research

The second most serious neurodegenerative disease is Parkinson’s disease (PD). Over the past several decades, a strong body of evidence suggests that PD can begin years before the hallmark clinical motor symptoms appear. Biomarkers for PD are urgently needed to differentiate between neurodegenerative disorders, screen novel therapeutics, and predict eventual clinical PD before the onset of symptoms. Some clinical evaluations and neuroimaging techniques have been developed in the last several years with some success in this area. Moreover, other strategies have been utilized to identify biochemical and genetic markers associated with PD leading to the examination of PD progression and pathogenesis in cerebrospinal fluid, blood, or saliva. Finally, interesting results are surfacing from preliminary studies using known PD-associated genetic mutations to assess potential premotor PD biomarkers. The current review highlights recent advances and underscores areas of potential advancement.

The validity and reliability of the Italian Olfactory Identification Test (IOIT) in healthy subjects and in Parkinson's disease patients

BACKGROUND

Olfactory function can be rapidly evaluated by means of standardized olfactory tests. Multiple-choice smell identification tests can be conditioned by cultural background. To investigate a new tool for detecting olfactory deficit in Italian subjects we developed a multiple-choice identification test prepared with odorants belonging to the Italian culture.

METHODS

The Italian Olfactory Identification Test (IOIT) was developed with 33 microencapsulated odorants with intensity of odors and headspace Gas Chromatography being tested. Test-retest reliability of the IOIT was evaluated. The IOIT was administered to 511 controls and 133 Parkinson's patients.

RESULTS

In healthy subjects the number of IOIT errors increased with age for both females (p < 0.0001) and males (p < 0.0001), while in the Parkinson's disease group the number of IOIT errors was significantly greater where compared to healthy subjects (p < 0.0001 in all age groups). The reference limits applied to all age groups revealed an IOIT sensitivity of 93% and a specificity of 99%. The test-retest reliability was excellent.

CONCLUSION

The IOIT is highly reliable, disposable, easy to administer, not fragile, and has a long shelf-life. All these features make the IOIT suitable for clinical use as well as for population screening and to discriminate Parkinson's patients from healthy subjects.

Screening for VPS35 mutations in Parkinson's disease

Recently 2 groups have independently identified a mutation in the gene 'vacuolar protein sorting 35 homolog' (VPS35 c.1858G>A; p.Asp620Asn) as a possible cause of autosomal dominant Parkinson's disease (PD). In order to assess the frequency of the reported mutation and to search for other possible disease-causing variants in this gene, we sequenced all 17 exons of VPS35 in 96 familial PD cases, and exon 15 (in which the reported mutation is found) in an additional 64 familial PD cases, 175 young-onset PD cases, and 262 sporadic, neuropathologically confirmed PD cases. We identified 1 individual with the p.Asp620Asn mutation and an autosomal dominant family history of PD. Subsequent follow-up of the family confirmed an affected sibling and cousin who also carried the same mutation. No other potentially disease-causing mutations were identified. We conclude that the VPS35 c.1858G>A mutation is an uncommon cause of familial Parkinson's disease in our population.

Extended extraocular phenotype of PROM1 mutation in kindreds with known autosomal dominant macular dystrophy

Mutations in prominin 1 (PROM1) have been shown to result in retinitis pigmentosa, macular degeneration and cone-rod dystrophy. Because of the putative role of PROM1 in hippocampal neurogenesis, we examined two kindreds with the same R373C PROM1 missense mutation using our established paradigm to study brain structure and function. As the protein encoded by PROM1, known as CD133, is used to identify stem/progenitor cells that can be found in peripheral blood and reflect endothelial reparatory mechanisms, other parameters were subsequently examined that included measures of vascular function, endothelial function and angiogenic capacity. We found that aspects of endothelial function assayed ex vivo were abnormal in patients with the R373C PROM1 mutation, with impaired adhesion capacity and higher levels of cellular damage. We also noted renal infections, haematuria and recurrent miscarriages possibly reflecting consequences of abnormal tubular modelling. Further studies are needed to confirm these findings.

Use of smell test identification in Parkinson's disease in Mexico: a matched case-control study

Smell tests can be useful in the differential diagnosis of Parkinson's disease (PD) but are affected by cultural factors. Currently there is no smell test tailored for the Mexican population but the brief smell identification test (B-SIT) was created as a cross-cultural SIT. We have created a translation of this test into Spanish adapted to the Mexican population and have applied it to 70 PD patients and 70 age- and gender-matched controls. The B-SIT differentiated PD and controls with 71.4% sensitivity and 85.7% specificity, when subjects were divided into two age groups.

Hyposmia in progressive supranuclear palsy

Previous studies suggested that olfaction is normal in progressive supranuclear palsy (PSP). We applied the University of Pennsylvania Smell Identification Test (UPSIT) to 36 patients with PSP who scored more than 18 on the Mini Mental State Examination (MMSE), 140 patients with nondemented Parkinson's disease (PD) and 126 controls. Mean UPSIT scores in PSP were lower than in controls (P < 0.001) but higher than in PD (P < 0.001) after adjusting for age, gender, and smoking history. For patients with PSP, UPSIT scores correlated with MMSE (r = 0.44, P = 0.006) but not disease duration (P = 0.6), motor subscale of the Unified Parkinson's Disease Rating Scale (P = 0.2), or Frontal Assessment Battery (P = 0.5). The brains of six of the patients with PSP were examined postmortem and all revealed neurofibrillary tangles and tau accumulation in the rhinencephalon, although only three had hyposmia. Further prospective studies including patients with early PSP and PSP-P with postmortem confirmation might help clarify if smell tests could be useful when the differential diagnosis lies between PD and PSP.