Browsing by Author "Fabbri, Margherita"
Now showing 1 - 10 of 20
Results Per Page
Sort Options
- Assessment of Parkinson’s disease medication state through automatic speech analysisPublication . Pompili, Anna; Solera-Urena, Rubén; Abad, Alberto; Cardoso, Rita; Guimarães, Isabel; Fabbri, Margherita; Martins, Isabel; Ferreira, JoaquimParkinson’s disease (PD) is a progressive degenerative disorder of the central nervous system characterized by motor and nonmotor symptoms. As the disease progresses, patients alternate periods in which motor symptoms are mitigated due to medication intake (ON state) and periods with motor complications (OFF state). The time that patients spend in the OFF condition is currently the main parameter employed to assess pharmacological interventions and to evaluate the efficacy of different active principles. In this work, we present a system that combines automatic speech processing and deep learning techniques to classify the medication state of PD patients by leveraging personal speech-based bio-markers. We devise a speakerdependent approach and investigate the relevance of different acoustic-prosodic feature sets. Results show an accuracy of 90.54% in a test task with mixed speech and an accuracy of 95.27% in a semi-spontaneous speech task. Overall, the experimental assessment shows the potentials of this approach towards the development of reliable, remote daily monitoring and scheduling of medication intake of PD patients.
- Assessment of Parkinson’s disease medication state through automatic speech analysisPublication . Pompili, Anna; Solera-Urena, Rubén; Abad, Alberto; Cardoso, Rita; Guimarães, Isabel; Fabbri, Margherita; Martins, Isabel P; Ferreira, JoaquimParkinson’s disease (PD) is a progressive degenerative disorder of the central nervous system characterized by motor and nonmotor symptoms. As the disease progresses, patients alternate periods in which motor symptoms are mitigated due to medication intake (ON state) and periods with motor complications (OFF state). The time that patients spend in the OFF condition is currently the main parameter employed to assess pharmacological interventions and to evaluate the efficacy of different active principles. In this work, we present a system that combines automatic speech processing and deep learning techniques to classify the medication state of PD patients by leveraging personal speech-based bio-markers. We devise a speakerdependent approach and investigate the relevance of different acoustic-prosodic feature sets. Results show an accuracy of 90.54% in a test task with mixed speech and an accuracy of 95.27% in a semi-spontaneous speech task. Overall, the experimental assessment shows the potentials of this approach towards the development of reliable, remote daily monitoring and scheduling of medication intake of PD patients.
- Clinical diagnostic criteria have a high impact on the frequency of dementia in late-stage Parkinson's diseasePublication . Sousa, Catarina Severiano e; Fabbri, Margherita; Godinho, Catarina; Simões, Rita; Chendo, Inês; Coelho, Miguel; Martins, Isabel Pavão; Ferreira, Joaquim J.Background: Cognitive impairment is a potential late feature of Parkinson's disease (PD). However, studies in patients with late-stage PD are lacking due to the particular characteristics of this population.
- Deep brain stimulation fine-tuning in Parkinson’s disease: short pulse width effect on speechPublication . Fabbri, Margherita; Natale, Federico; Artusi, Carlo Alberto; Romagnolo, Alberto; Bozzali, Marco; Giulietti, Giovanni; Guimarães, Isabel; Rizonni, Mario Georgio; Accornero, Anna; Lopiano, Leonardo; Zibetti, MaurizioBackground: subthalamic nucleus deep brain stimulation (STN-DBS) may have a detrimental effect on speech in Parkinson’s disease (PD) patients and new stimulation technologies may help in addressing this issue. Objective: to evaluate the STN-DBS acute effect of 30 μs pulse width (30PW) versus conventional 60 μs PW (60PW) on speech and identify the core features of voice modified by 30PW. Methods: seven STN-DBS treated PD patients participated into a pilot cross-sectional study. Motor and speech performances were tested by means of both automatic analysis and blinded clinical evaluations in four stimulation conditions: 30PW and 60PW both at the usual amplitude and at an amplitude just below the threshold for stimulation-related side effects. Results: at the threshold amplitude, 30PW stimulation improved speech intelligibility for both words (p = 0.02) and sentences (p = 0.04), without worsening motor performance. A lower but not statistically significant voice variability and instability and percentage of stuttering disfluencies was also observed. The beneficial effect of 30PW detected by automatic analysis, was confirmed by patients’ perception. Conclusions: STN-DBS treated patients experiencing low speech intelligibility may benefit from a 30PW stimulation trial at a higher amplitude. Deep characterization of PD speech profiles may help in a better application of recent DBS hardware advances.
- Disability in activities of daily living and severity of dyskinesias determine the handicap of Parkinson´s disease patients in advanced stage selected to dbsPublication . Coelho, Miguel; Abreu, Daisy; Correia-Guedes, Leonor; Lobo, Patrícia Pita; Fabbri, Margherita; Godinho, Catarina; Domingos, Josefa; Albuquerque, Luisa; Freitas, Vanda; Pereira, João Miguel; Cattoni, Begona; Carvalho, Herculano; Reimão, Sofia; Rosa, Mário M.; Ferreira, António Gonçalves; Ferreira, Joaquim J."BACKGROUND: There is scarce data on the level of handicap in Parkinson's disease (PD) and none in advanced stage PD. OBJECTIVE: To assess the handicap in advanced stage PD patients with disabling levodopa-induced motor complications selected to deep brain stimulation (DBS). METHODS: Data was prospectively recorded during routine evaluation for DBS. Handicap was measured using London Handicap Scale (LHS) (0 = maximal handicap; 1 = no handicap). Disease severity was evaluated using the Hoehn & Yahr scale and the UPDRS/MDS-UPDRS, during off and on after a supra-maximal dose of levodopa. Schwab and England Scale (S&E) was scored in off and on. Dyskinesias were scored using the modified Abnormal Involuntary Movement Scale (mAIMS). Results concern cross-sectional assessment before DBS. RESULTS: 100 PD patients (mean age 61 (±7.6); mean disease duration 12.20 (±4.6) years) were included. Median score of motor MDS-UPDRS was 54 in off and 25 in on. Mean total LHS score was 0.56 (±0.14). Patients were handicapped in several domains with a wide range of severity. Physical Independence and Social Integration were the most affected domains. Determinants of total LHS score were MDS-UPDRS part II off (β= -0.271; p = 0.020), S&E on (β= 0.264; p = 0.005) and off (β= 0.226; p = 0.020), and mAIMS on (β= -0.183; p = 0.042) scores (R2 = 29.6%). CONCLUSIONS: We were able to use handicap to measure overall health condition in advanced stage PD. Patients were moderately to highly handicapped and this was strongly determined by disability in ADL and dyskinesias. Change in handicap may be a good patient-centred outcome to assess efficiency of DBS."
- Dysphagia predicts poor outcome in late-stage Parkinson's diseasePublication . Fabbri, Margherita; Coelho, Miguel; Abreu, Daisy; Guedes, Leonor Correia; Rosa, Mario M.; Godinho, Catarina; Cardoso, Rita; Guimarães, Isabel; Antonini, Angelo; Zibetti, Maurizio; Lopiano, Leonardo; Ferreira, Joaquim J."BACKGROUND: Few data exist on the rate of clinical progression for Parkinson's disease (PD) patients who have entered a late stage of the disease. OBJECTIVE: Study the clinical progression of a late-stage PD (LSPD) population over one year follow-up. METHODS: 50 LSPD patients (Schwab and England ADL Scale <50 or Hoehn Yahr Stage >3 in MED ON) underwent an extensive clinical assessment at baseline and after one year and an acute levodopa test at baseline. RESULTS: Mean age of LSPD patients (female 46%) was 77.5 ± 5.9 years and mean disease duration was 15.5 ± 6.5 years. At baseline, 76% had levodopa-induced motor complications (MC), usually non-troublesome, 68% were demented, 54% had psychosis and 68% depression. Caregiver distress was high. l-dopa responsiveness was mild (18% ± 12 of improvement on MDS-UPDRS-III). After one-year, 20% of the patients were dead, institutionalized or HY 5. MDS-UPDRS-motor mean score worsened 7.2 ± 10.3 points although there was heterogeneity between patients, and there was a global worsening of non-motor symptoms, mostly in cognition/mood, urinary and gastrointestinal domains. Nevertheless, MC improved despite similar levodopa equivalent dose. Functional independence and quality of life worsened. Dysphagia severity at baseline predicted a poor outcome (death, institutionalization or HY 5) (Hazard ratio 2.3, 95% CI 1.12-4.4; p = 0.01), whereas magnitude of l-dopa response of LSPD patients did not. CONCLUSIONS: LSPD patients still present a significant, although heterogeneous, motor and non-motor progression over 1 year. Dysphagia severity predicts the occurrence of additional disease severity milestones and its management must be prioritized."
- Dysphagia predicts poor outcome in late-stage Parkinson's diseasePublication . Fabbri, Margherita; Coelho, Miguel; Abreu, Daisy; Guedes, Leonor Correia; Rosa, Mario M; Godinho, Catarina; Cardoso, Rita; Guimarães, Isabel; Antonini, Angelo; Zibetti, Maurizio; Lopiano, Leonardo; Ferreira, Joaquim J
- High frequency of psychosis in late-stage Parkinson’s diseasePublication . Chendo, Inês; Fabbri, Margherita; Godinho, Catarina; Simões, Rita Moiron; Sousa, Catarina Severiano; Coelho, Miguel; Voon, Valerie; Ferreira, Joaquim J.Background Psychosis is a frequent non-motor symptom in Parkinson’s disease (PD). Estimates of the frequency of Parkinsońs disease psychosis (PDP) vary widely. Knowledge about the frequency and phenomenology of psychosis in late-stage (LS) PD patients is limited. This study aimed to determine the frequency of psychosis in LSPD patients through clinical diagnostic interview (CDI) (gold standard), according to NINDS/NIMH diagnostic criteria for PDP. The secondary objectives were to characterize the phenomenology, to test selected instruments and assess their adequacy in comparison to CDI, and to assess the psychiatric comorbidities.
- Is lowering stimulation frequency a feasible option for subthalamic deep brain stimulation in Parkinson's disease patients with dysarthria?Publication . Fabbri, Margherita; Zibetti, Maurizio; Ferrero, Giulia; Accornero, Anna; Guimarães, Isabel; Rizzone, Mario Giorgio; Romagnolo, Alberto; Ferreira, Joaquim J; Lopiano, LeonardoBackground: The long-term effect of subthalamic nucleus deep brain stimulation (STN-DBS) on dysarthria can be detrimental. A transient beneficial effect of low-frequency stimulation (LFS) has been reported. Objective: to investigate if the magnitude of dysarthria could predict the effect of LFS on speech in STN-DBS PD patients and to verify whether the benefit is maintained over time. Methods: a cohort study, comparing 10 PD patients with severe speech impairment (MDS-UPDRS item 3.1≥3) with 10 PD patients with mild speech impairment (MDS-UPDRS item 3.1≤2), all submitted to STN-DBS. Patients were tested in: MED OFF/STIM OFF, MED OFF/STIM ON (130 Hz, high frequency stimulation [HFS]), MED OFF/STIM ON (60 Hz - LFS) and MED ON with both HFS and LFS. The following was assessed in all conditions: voice (fundamental frequency and jitter), speech (articulatory diadochokinesis [DDK], pitch variability, rate and intelligibility) and motor performance (MDS-UPDRS-III). Results: LFS compared to no stimulation and HFS, in the absence of L-dopa effect, significantly improved DDK and speech intelligibility for sentence, among patients with severe speech impairment. During the L-dopa effect, comparing LFS vs. HFS, there was a significant improvement of speech intelligibility in both groups. Five patients with severe speech impairment opted to maintain LFS. After six months, speech benefit was maintained but treatment adjustments were required. Conclusions: LFS may offer both an immediate and long-lasting improvement of speech in a subgroup of STNDBS patients with severe speech impairment during HFS. Nevertheless, its effect on motor symptoms may not be preserved over time.
- Main reasons for rejection of deep brain stimulation surgery in candidates with Parkinson DiseasePublication . Sousa, Teresa Almeida e; Ferret-Sena, Véronique; Coelho, Miguel; Correia-Guedes, Leonor; Fabbri, Margherita; Rosa, Mário Miguel; Godinho, Catarina