Fully automated cognitive screening tool based on assessment of speech and language.

INTRODUCTION: Recent years have seen an almost sevenfold rise in referrals to specialist memory clinics. This has been associated with an increased proportion of patients referred with functional cognitive disorder (FCD), that is, non-progressive cognitive complaints. These patients are likely to benefit from a range of interventions (eg, psychotherapy) distinct from the requirements of patients with neurodegenerative cognitive disorders. We have developed a fully automated system, 'CognoSpeak', which enables risk stratification at the primary-secondary care interface and ongoing monitoring of patients with memory concerns. METHODS: We recruited 15 participants to each of four groups: Alzheimer's disease (AD), mild cognitive impairment (MCI), FCD and healthy controls. Participants responded to 12 questions posed by a computer-presented talking head. Automatic analysis of the audio and speech data involved speaker segmentation, automatic speech recognition and machine learning classification. RESULTS: CognoSpeak could distinguish between participants in the AD or MCI groups and those in the FCD or healthy control groups with a sensitivity of 86.7%. Patients with MCI were identified with a sensitivity of 80%. DISCUSSION: Our fully automated system achieved levels of accuracy comparable to currently available, manually administered assessments. Greater accuracy should be achievable through further system training with a greater number of users, the inclusion of verbal fluency tasks and repeat assessments. The current data supports CognoSpeak's promise as a screening and monitoring tool for patients with MCI. Pending confirmation of these findings, it may allow clinicians to offer patients at low risk of dementia earlier reassurance and relieve pressures on specialist memory services.

Cognition and behaviour in frontotemporal dementia with and without amyotrophic lateral sclerosis.

OBJECTIVE: The precise relationship between frontotemporal dementia (FTD) and amyotrophic lateral sclerosis (ALS) is incompletely understood. The association has been described as a continuum, yet data suggest that this may be an oversimplification. Direct comparisons between patients who have behavioural variant FTD (bvFTD) with and without ALS are rare. This prospective comparative study aimed to determine whether there are phenotypic differences in cognition and behaviour between patients with FTD-ALS and bvFTD alone. METHODS: Patients with bvFTD or FTD-ALS and healthy controls underwent neuropsychological testing, focusing on language, executive functions and social cognition. Behavioural change was measured through caregiver interview. Blood samples were screened for known FTD genes. RESULTS: 23 bvFTD, 20 FTD-ALS and 30 controls participated. On cognitive tests, highly significant differences were elicited between patients and controls, confirming the tests' sensitivities to FTD. bvFTD and FTD-ALS groups performed similarly, although with slightly greater difficulty in patients with ALS-FTD on category fluency and a sentence-ordering task that assesses grammar production. Patients with bvFTD demonstrated more widespread behavioural change, with more frequent disinhibition, impulsivity, loss of empathy and repetitive behaviours. Behaviour in FTD-ALS was dominated by apathy. The C9ORF72 repeat expansion was associated with poorer performance on language-related tasks. CONCLUSIONS: Differences were elicited in cognition and behaviour between bvFTD and FTD-ALS, and patients carrying the C9ORF72 repeat expansion. The findings, which raise the possibility of phenotypic variation between bvFTD and FTD-ALS, have clinical implications for early detection of FTD-ALS and theoretical implications for the nature of the relationship between FTD and ALS.

Baseline characteristics, analysis plan and report on feasibility for the Prevention Of Decline in Cognition After Stroke Trial (PODCAST).

BACKGROUND: A common complication after stroke is development of cognitive impairment and dementia. However, effective strategies for reducing the risk of developing these problems remain undefined. Potential strategies include intensive lowering of blood pressure (BP) and/or lipids. This paper summarises the baseline characteristics, statistical analysis plan and feasibility of a randomised control trial of blood pressure and lipid lowering in patients post-stroke with the primary objective of reducing cognitive impairment and dementia. METHODS: The Prevention Of Decline in Cognition After Stroke Trial (PODCAST) was a multi-centre prospective randomised open-label blinded-endpoint controlled partial-factorial internal pilot trial running in secondary and primary care. Participants without dementia were enrolled 3-7 months post ischaemic stroke or spontaneous intracerebral haemorrhage, and randomised to intensive versus guideline BP lowering (target systolic BP <125 mmHg versus <140 mmHg); patients with ischaemic stroke were also randomised to intensive or guideline lipid lowering (target LDL cholesterol <1.4 mmol/L versus <3 mmol/L). The primary outcome was the Addenbrooke's Cognitive Examination-Revised; a key secondary outcome was to assess feasibility of performing a large trial of one or both interventions. Data are number (%) or mean (standard deviation). The trial was planned to last for 8 years with follow-up between 1 and 8 years. The plan for reporting the main results is included as Additional file 2. RESULTS: 83 patients (of a planned 600) were recruited from 19 UK sites between 7 October 2010 and 31 January 2014. Delays, due to difficulties in the provision of excess treatment costs and to complexity of follow-up, led to few centres taking part and a much lower recruitment rate than planned. Patient characteristics at baseline were: age 74 (SD 7) years, male 64 (77 %), index stroke ischaemic 77 (93 %), stroke onset to randomisation 4.5 [SD 1.3] months, Addenbrooke's Cognitive Examination-Revised 86 (of 100, SD 8), Montreal Cognitive Assessment 24 (of 30, SD 3), BP 147/82 (SD 19/11) mmHg, total cholesterol 4.0 (SD 0.8) mmol/L and LDL cholesterol 2.0 (SD 0.7) mmol/L, modified Rankin Scale 1.1 (SD 0.8). CONCLUSION: Limited recruitment suggests that a large trial is not feasible using the current protocol. The effects of the interventions on BP, lipids, and cognition will be reported in the main publication. TRIAL REGISTRATION: ISRCTN85562386 registered on 23 September 2009.