This is a non-invasive study investigating whether EEG, which is used in routine clinical care can be useful in diagnosing dementia. Novel in house developed software will analyse EEG data for this purpose from people with Alzheimer's disease and healthy volunteers.

Participant recruitment

15 people with early AD will be recruited from the Sheffield Teaching Hospital and Sheffield Health and Social Care NHS Trusts memory clinics. The STH clinic is neurology led and sees working age (<65 years old) patients while the SHSC clinic sees an older (>65 years old) population. Over 1300 new assessments are conducted every year. Approximately 65-70% are diagnosed with dementia.

Investigators will choose patients with typical neuropsychological profiles of AD. All patients will have previously been recruited to the Virtual Physiological Human: Dementia Research enabled by IT. VPH-DARE@IT is recruiting 30-40 participants in each of the three following groups (mild AD, MCI and normal controls) and thus our participants will be drawn from this group and all will have pre-existing high-resolution 3D structural and volumetric assessment plus resting-state fMRI scans. This resting-state fMRI data will be analysed to examine the DMN, allowing us to directly compare measures of brain-connectivity between EEG and fMRI and contrast the discriminatory ability of the two methodologies.

10 healthy controls will be aged matched and recruited either from Neurology clinics or be spouses of study participants. The investigators will endeavour to recruit normal controls who have undergone fMRI as part of the academic neuroradiology departments control population for VPH-DARE@IT and other studies.

People with AD and 10 healthy controls will have a repeat EEG to assess the reproducibility and robustness of the EEG measures.

EEG is normally very well tolerated and as listed in the background has been used in many people with dementia. The Clinical Neurophysiology department is very experienced in undertaking EEG in all patient populations from neonates to the very elderly.

Consent Participants will be identified by the clinicians working in the memory clinic or through posters in memory clinic and at other University sites to attract normal volunteers. A Participation Information Sheet (PIS) will be given to them either directly or via post. Participants will have sufficient time to read the PIS and be able to ask questions. The PIS will inform the participant of the benefits and potential risks involved in taking part in this study. They will be made aware of the possibility that incidental changes are rarely detected. If this is seen they would be informed and we would arrange appropriate management. The person will be consented when they come in for the EEG recordings. Investigators will explain to all participants that they will be able to leave the study at any time including during the EEG recording and in those that we plan to repeat the EEG can decline the 2nd test. The Clinical Department of Neurophysiology has experience in performing EEG in all ages including those with dementia and cognitive impairment. The person with dementia's care-giver or family member can come to the department and be in the room for the entire time the EEG recording takes place.

Methodology EEG data capture

• Participants will attend the department of clinical neurophysiology and a member of the research team will take consent. Two experienced EEG technicians will place the electrodes and explain the procedure. We will record resting state EEG. Participants will be told to lie still with eyes closed. It will be explained that although their eyes are closed they must remain awake. Participants will be continuously monitored and if they become drowsy will be re-stimulated and the time of drowsiness will be recorded. This instruction and monitoring is the same as that used in previous EEG and fMRI studies looking at resting state paradigms.
• The 10-10 International system of electrode placement will be used for the scalp EEG recordings with a multichannel high sampling rate amplifier (2 KHz). This will provide standard 10/20 electrode positions (F3, F4, F7, F8, C3, C4, Fz, Cz, T3, T4, T5, T6, P3, P4, O1, O2, and Oz electrodes) which will be initially used to detect the more significant spatiotemporal measurements/phenomena while offering the opportunity to elaborate on the spatial and frequency resolution of our findings at a later stage. Peri-ocular EOG electrode data will also be recorded to detect eye movement related slow wave activity. Investigators will therefore be able to avoid spurious estimates of slow wave activity in the anterior EEG regions.
• Artefact free epochs within suitable (wakeful no task eyes closed state) will be selected and exported for further quantitative EEG analysis.
• The exported data will be stored on an external encrypted drive.
• All the analyses will be performed on custom made in-house software on a high tech computer located within secure offices locked within Dr Sarrigannis's office housed in the department of clinical neurophysiology.

EEG Data analyses • The error reduction ratio-causality method will be used to calculate levels of synchronisation. The standard EEG 10-20 electrodes, including all possible contiguous electrode combinations in pairs of two (e.g. right frontocentral versus parietoccipital) will be used in our pilot exploration to assess the 95% confidence interval levels of synchronisation and determine if they allow robust separation between healthy volunteers and patients with Alzheimer's Disease.

Other Data

• Qualitative Data collection: Investigators will collect questionnaires on the tolerability of undergoing EEG examination (please find enclosed questionnaire).

Ethical consideration Modern neuro-imaging using MRI is highly sensitive and can detect incidental findings. Our novel EEG analyses also have the potential to detect incidental findings. It is well known that spikes during an inter-ictal EEG have limited clinical significance. However they are seen more frequently in people who have a liability, or lowered threshold, for epilepsy. If any concern is raised about the EEG findings the participant will be reviewed by Dr Blackburn and if appropriate referred onto another neurological colleague. Any incidental findings detected by MRI will be referred to an NHS neurologist through the VPH-DARE@IT project researchers. The Patient Information Sheet and consent will make participants aware of these risks before they take part in the study.

Statistics This pilot project will explore EEG data to investigate the frequency and time domain measurement that would be most appropriate for clinical use. Investigators do not know whether this will involve linear or non-linear interaction and thus we cannot determine which is the most appropriate statistical test. The analysis of similar EEG data sets using parametric methods has explored this in more detail.

Power calculations As this is a proof-of-concept, pilot study to examine the robustness of a novel methodology, it is inappropriate to conduct a formal power calculation. Our initial findings will determine whether our EEG analyses detects functional connectivity correlates of the synchronisation of anterior and posterior networks identified by fMRI BOLD signal during resting state scanning. Investigators will also explore if our data is comparable to previously published quantitative EEG on people with AD. Investigators will explore which measures from our data can discriminate people with Alzheimer's disease from healthy controls. These initial results will give us likely effect sizes and allow us to conduct more accurate power calculations than are presently possible. This will be used to determine the sample size of a larger study of people with dementia.

Inclusion Criteria:

• Clinical diagnosis of possible or probable AD (according to revised diagnostic criteria (McKhann et al., 2011, Dubois et al., 2007)

Exclusion Criteria:

• Significant vascular change on neuro-imaging (Fazekas grading of white matter change >2) (Fazekas et al., 2002).
• Clinical history of stroke, Parkinsonism, epilepsy or other neurological disorder. (e.g. severe auditory, visual, language or motor impairment)
• Moderate to severe dementia (defined by research team but we wouldn't expect those with a mini mental state exam [MMSE] ≤ 18 to be able to participate)
• Severe current mental health disorder (depression, psychosis).
• Patients on sedating or antiepileptic medication

Clinical diagnosis of possible or probable AD (according to revised diagnostic criteria (McKhann et al., 2011, Dubois et al., 2007)