Towards a computerized estimation of visual complexity in images: Data to assess the association of computed visual complexity features to human responses in visual tasks.

Artificial vision has been extensively studied in the mathematical and computational Sciences. Concurrently, psychological studies attempt to describe visual cognition and the complexity of visual tasks as perceived by humans. The methods and the definitions of vision used by these two disciplines are disjointed. Particularly, an explanation of computer vision performance by human-perceived attributes, if attempted, can only be inferred. This article describes a dataset collected to explore the association between computer-extracted visual attributes and human-perceived attributes in the context of cognitive tasks. The data was acquired from a cohort of 406 subjects, ages 40-90, in the presence of a healthcare professional who assessed that the subjects had no cognitive or motor disorder. The subjects performed computerized cognitive tests which entailed tasks of recognition or recall of an image in a set of three images, presented on the computer screen. The images were simple black and white abstract square shapes. The latencies of the subjects' responses, by keyboard key press, to each task were logged. The data contains 3 parts: the images presented in each task, described by binary vectors for black and white coding, a response time logged for each task and the subjects' age, gender, and computer proficiency. A preliminary comparison of computationally-extracted complexity features and subjects' performance is provided in the article entitled "Linking computerized and perceived attributes of visual complexity" [1].

SARS-CoV-2 Detection From Voice.

Automated voice-based detection of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) could facilitate the screening for COVID19. A dataset of cellular phone recordings from 88 subjects was recently collected. The dataset included vocal utterances, speech and coughs that were self-recorded by the subjects in either hospitals or isolation sites. All subjects underwent nasopharyngeal swabbing at the time of recording and were labelled as SARS-CoV-2 positives or negative controls. The present study harnessed deep machine learning and speech processing to detect the SARS-CoV-2 positives. A three-stage architecture was implemented. A self-supervised attention-based transformer generated embeddings from the audio inputs. Recurrent neural networks were used to produce specialized sub-models for the SARS-CoV-2 classification. An ensemble stacking fused the predictions of the sub-models. Pre-training, bootstrapping and regularization techniques were used to prevent overfitting. A recall of 78% and a probability of false alarm (PFA) of 41% were measured on a test set of 57 recording sessions. A leave-one-speaker-out cross validation on 292 recording sessions yielded a recall of 78% and a PFA of 30%. These preliminary results imply a feasibility for COVID19 screening using voice.

The response of peripheral microcirculation to gravity-induced changes.

BACKGROUND: The peripheral microcirculation supplies fresh blood to the small blood vessels, providing oxygen and nutrients to the tissues, removing waste, and maintaining normal homeostatic conditions. The goal of this study was to characterize the response of the peripheral microcirculation, in terms of blood flow and tissue oxygenation variables, to gravity-induced changes. METHODS: The study included 20 healthy volunteers and the experiment involved monitoring central and peripheral variables with the right hand positioned at different heights. These positions correspond to various gravitational levels. Peripheral perfusion and oxygenation were monitored using a laser Doppler flowmeter, photoplethysmograph, and transcutaneous oxygen tension monitor. Local blood pressure and respiration rate were also measured. FINDINGS: At the heart level, tissue oxygenation displayed a nadir, while capillary flow displayed a peak. Similar but weaker changes were observed at the control hand. In contrast, the photoplethysmograph's amplitude strongly decreased upon reducing the arm position below heart level. Both systolic and diastolic pressures decreased linearly between the highest to lowest arm position. INTERPRETATION: The results may reflect peripheral compensation mechanisms, as well as an interaction between the central and peripheral cardiovascular systems, in response to local changes in blood pressure. The observed dependence of the oxygenation pattern on height could lead to important new insights for the diagnosis and treatment of problems in the regulation of tissue perfusion.

A real-time phoneme counting algorithm and application for speech rate monitoring.

Adults who stutter can learn to control and improve their speech fluency by modifying their speaking rate. Existing speech therapy technologies can assist this practice by monitoring speaking rate and providing feedback to the patient, but cannot provide an accurate, quantitative measurement of speaking rate. Moreover, most technologies are too complex and costly to be used for home practice. We developed an algorithm and a smartphone application that monitor a patient's speaking rate in real time and provide user-friendly feedback to both patient and therapist. Our speaking rate computation is performed by a phoneme counting algorithm which implements spectral transition measure extraction to estimate phoneme boundaries. The algorithm is implemented in real time in a mobile application that presents its results in a user-friendly interface. The application incorporates two modes: one provides the patient with visual feedback of his/her speech rate for self-practice and another provides the speech therapist with recordings, speech rate analysis and tools to manage the patient's practice. The algorithm's phoneme counting accuracy was validated on ten healthy subjects who read a paragraph at slow, normal and fast paces, and was compared to manual counting of speech experts. Test-retest and intra-counter reliability were assessed. Preliminary results indicate differences of -4% to 11% between automatic and human phoneme counting. Differences were largest for slow speech. The application can thus provide reliable, user-friendly, real-time feedback for speaking rate control practice.

The coupling between peripheral microcirculation and slow breathing.

Vasomotion (rhythmic changes in arteriolar diameter) is believed to enhance tissue perfusion at low oxygenation levels. We hypothesized that slow breathing and vasomotion may correlate temporally ("coupling"), especially at low oxygenation levels. We paced down spontaneous breathing to about 5 or 6breaths/min in 14 healthy subjects using device-guided breathing (DGB), and continuously monitored respiration, transcutaneous oxygen pressure ("oxygenation"), and skin capillary blood flow ("microflow") using a laser Doppler flowmeter. The coupling was expressed by cross-correlation calculated in 1-min time windows. Our main results illustrated that: (1) coupling increased gradually upon slowing breathing down in a subgroup, in which initial oxygenation was lower than a threshold of 30mmHg (0.3±0.2 vs. 0.07±0.2, P<10-6); (2) during DGB changes in oxygenation elicited opposite (relative) changes in microflow, with 4-fold higher sensitivity for low initial oxygenation relative to high (regression slope -0.094±0.010mmHg-1 vs. -0.020±0.002mmHg-1, P<10-6); (3) at low initial oxygenation, we observed larger coupling and (relative) microflow changes in younger subjects, and greater oxygenation changes in females (P<10-6 for all); (4) pulse pressure changes from before to after DGB were reduced by increased oxygenation changes during DGB (-5.5±7.4mmHg, r=-0.73, P<0.001). In conclusion, the present methodology can provide the variation trend of respiration-vasomotion coupling during DGB that may characterize microcirculation behavior at tissue oxygenation below a measurable threshold. The potential association of these trends and thresholds with pathologies or specific conditions of the cardiopulmonary system, and the possible role played by the neural sympathetic activity in that coupling, deserve further studies.

Computer-based, personalized cognitive training versus classical computer games: a randomized double-blind prospective trial of cognitive stimulation.

BACKGROUND: Many studies have suggested that cognitive training can result in cognitive gains in healthy older adults. We investigated whether personalized computerized cognitive training provides greater benefits than those obtained by playing conventional computer games. METHODS: This was a randomized double-blind interventional study. Self-referred healthy older adults (n = 155, 68 ± 7 years old) were assigned to either a personalized, computerized cognitive training or to a computer games group. Cognitive performance was assessed at baseline and after 3 months by a neuropsychological assessment battery. Differences in cognitive performance scores between and within groups were evaluated using mixed effects models in 2 approaches: adherence only (AO; n = 121) and intention to treat (ITT; n = 155). RESULTS: Both groups improved in cognitive performance. The improvement in the personalized cognitive training group was significant (p < 0.03, AO and ITT approaches) in all 8 cognitive domains. However, in the computer games group it was significant (p < 0.05) in only 4 (AO) or 6 domains (ITT). In the AO analysis, personalized cognitive training was significantly more effective than playing games in improving visuospatial working memory (p = 0.0001), visuospatial learning (p = 0.0012) and focused attention (p = 0.0019). CONCLUSIONS: Personalized, computerized cognitive training appears to be more effective than computer games in improving cognitive performance in healthy older adults. Further studies are needed to evaluate the ecological validity of these findings.

Computerized methods in the assessment and prediction of dementia.

Computerized administration of neuropsychological tests can be an objective, sensitive and efficient way to screen for and monitor cognitive changes in the elderly. However, current computer software still suffers from limitations in both the administration of those tests and the interpretation of their results, which might severely hamper their usability. In this paper qualitative aspects of current methods and their use in the prediction of dementia are discussed, guidelines for correct design and usage of computerized methods are suggested and a solution that overcomes several of the methodological limitations is proposed.

Computerized diagnosis of mild cognitive impairment.

BACKGROUND: We previously described software that we have developed for use in the evaluation of mild cognitive impairment (MCI). Our previous study included an aged nondemented population with memory complaints (n = 41) that was relatively homogenous in terms of education, clinical history, neurological examination, and Mini-Mental Status Examination (MMSE) scores. Performance patterns in the computerized tests separated the subjects into two groups, and we hypothesized that one group might have had incipient dementia. METHODS/RESULTS: In the present study we report a follow-up of 35 of the subjects 2 years later. Eight subjects who were thought to have incipient dementia at baseline could be evaluated in the follow-up, and six of them have deteriorated according to both MMSE and neurologists' evaluations and have now fulfilled clinical diagnostic criteria of dementia. The other two deteriorated only according to their computer performance. Of the 27 remaining subjects, only one now fulfilled clinical diagnostic criteria for dementia, although the present computerized examinations identified 10 subjects whose performance has deteriorated compared with the previous session. CONCLUSION: The follow-up examination thus supported our hypothesis that human-computer interaction features can contribute to the detection of incipient dementia.

Human-computer interaction in the administration and analysis of neuropsychological tests.

We developed a computerized neuropsychological assessment software, which employs innovative features in both the presentation of the tests and the interpretation of the subjects' performance. The usability features of the software enabled elderly subjects with no computer experience to undergo the cognitive tests, without a test administrator being present. New features extracted from reaction times allowed for enhancement of the performance analysis of the tests. The software was validated using a large group of controls and then administered to a group of subjects with age related memory complaints. The results indicate an improved sensitivity of two of the computerized tests compared with the mini-mental state examination, which enabled to separate two groups of subjects from a population that was relatively homogenous based on clinical history, neurological examination and the mini-mental state examination.

Medial olivocochlear efferent reflex in humans: otoacoustic emission (OAE) measurement issues and the advantages of stimulus frequency OAEs.

Otoacoustic emissions (OAEs) are useful for studying medial olivocochlear (MOC) efferents, but several unresolved methodological issues cloud the interpretation of the data they produce. Most efferent assays use a "probe stimulus" to produce an OAE and an "elicitor stimulus" to evoke efferent activity and thereby change the OAE. However, little attention has been given to whether the probe stimulus itself elicits efferent activity. In addition, most studies use only contralateral ( re the probe) elicitors and do not include measurements to rule out middle-ear muscle (MEM) contractions. Here we describe methods to deal with these problems and present a new efferent assay based on stimulus frequency OAEs (SFOAEs) that incorporates these methods. By using a postelicitor window, we make measurements in individual subjects of efferent effects from contralateral, ipsilateral, and bilateral elicitors. Using our SFOAE assay, we demonstrate that commonly used probe sounds (clicks, tone pips, and tone pairs) elicit efferent activity, by themselves. Thus, results of efferent assays using these probe stimuli can be confounded by unwanted efferent activation. In contrast, the single 40 dB SPL tone used as the probe sound for SFOAE-based measurements evoked little or no efferent activity. Since they evoke efferent activation, clicks, tone pips, and tone pairs can be used in an adaptation efferent assay, but such paradigms are limited in measurement scope compared to paradigms that separate probe and elicitor stimuli. Finally, we describe tests to distinguish middle-ear muscle (MEM) effects from MOC effects for a number of OAE assays and show results from SFOAE-based tests. The SFOAE assay used in this study provides a sensitive, flexible, frequency-specific assay of medial efferent activation that uses a low-level probe sound that elicits little or no efferent activity, and thus provides results that can be interpreted without the confound of unintended efferent activation.