Detecting the patient's need for help with machine learning based on expressions.

BACKGROUND: Developing machine learning models to support health analytics requires increased understanding about statistical properties of self-rated expression statements used in health-related communication and decision making. To address this, our current research analyzes self-rated expression statements concerning the coronavirus COVID-19 epidemic and with a new methodology identifies how statistically significant differences between groups of respondents can be linked to machine learning results. METHODS: A quantitative cross-sectional study gathering the "need for help" ratings for twenty health-related expression statements concerning the coronavirus epidemic on an 11-point Likert scale, and nine answers about the person's health and wellbeing, sex and age. The study involved online respondents between 30 May and 3 August 2020 recruited from Finnish patient and disabled people's organizations, other health-related organizations and professionals, and educational institutions (n = 673). We propose and experimentally motivate a new methodology of influence analysis concerning machine learning to be applied for evaluating how machine learning results depend on and are influenced by various properties of the data which are identified with traditional statistical methods. RESULTS: We found statistically significant Kendall rank-correlations and high cosine similarity values between various health-related expression statement pairs concerning the "need for help" ratings and a background question pair. With tests of Wilcoxon rank-sum, Kruskal-Wallis and one-way analysis of variance (ANOVA) between groups we identified statistically significant rating differences for several health-related expression statements in respect to groupings based on the answer values of background questions, such as the ratings of suspecting to have the coronavirus infection and having it depending on the estimated health condition, quality of life and sex. Our new methodology enabled us to identify how statistically significant rating differences were linked to machine learning results thus helping to develop better human-understandable machine learning models. CONCLUSIONS: The self-rated "need for help" concerning health-related expression statements differs statistically significantly depending on the person's background information, such as his/her estimated health condition, quality of life and sex. With our new methodology statistically significant rating differences can be linked to machine learning results thus enabling to develop better machine learning to identify, interpret and address the patient's needs for well-personalized care.

Gas7 Is a Novel Dendritic Spine Initiation Factor.

Brain stores new information by modifying connections between neurons. When new information is learnt, a group of neurons gets activated and they are connected to each other via synapses. Dendritic spines are protrusions along neuronal dendrites where excitatory synapses are located. Dendritic spines are the first structures to protrude out from the dendrite to reach out to other neurons and establish a new connection. Thus, it is expected that neuronal activity enhances spine initiation. However, the molecular mechanisms linking neuronal activity to spine initiation are poorly known. Membrane binding BAR domain proteins are involved in spine initiation, but it is not known whether neuronal activity affects BAR domain proteins. Here, we used bicuculline treatment to activate excitatory neurons in organotypic hippocampal slices. With this experimental setup, we identified F-BAR domain containing growth arrest-specific protein (Gas7) as a novel spine initiation factor responding to neuron activity. Upon bicuculline addition, Gas7 clustered to create spine initiation hotspots, thus increasing the probability to form new spines in activated neurons. Gas7 clustering and localization was dependent on PI3-kinase (PI3K) activity and intact F-BAR domain. Gas7 overexpression enhanced N-WASP localization to clusters as well as it increased the clustering of actin. Arp2/3 complex was required for normal Gas7-induced actin clustering. Gas7 overexpression increased and knock-down decreased spine density in hippocampal pyramidal neurons. Taken together, we suggest that Gas7 creates platforms under the dendritic plasma membrane which facilitate spine initiation. These platforms grow on neuronal activation, increasing the probability of making new spines and new connections between active neurons. As such, we identified a novel molecular mechanism to link neuronal activity to the formation of new connections between neurons.

Measuring Properties of the Membrane Periodic Skeleton of the Axon Initial Segment using 3D-Structured Illumination Microscopy (3D-SIM).

The axon initial segment (AIS) is the site at which action potentials initiate and constitutes a transport filter and diffusion barrier that contribute to the maintenance of neuronal polarity by sorting somato-dendritic cargo. A membrane periodic skeleton (MPS) comprising periodic actin rings provides a scaffold for anchoring various AIS proteins, including structural proteins and different ion channels. Although recent proteomic approaches have identified a considerable number of novel AIS components, details of the structure of the MPS and the roles of its individual components are lacking. The distance between individual actin rings in the MPS (~190 nm) necessitates the employment of super-resolution microscopy techniques to resolve the structural details of the MPS. This protocol describes a method for using cultured rat hippocampal neurons to examine the precise localization of an AIS protein in the MPS relative to sub-membranous actin rings using 3D-structured illumination microscopy (3D-SIM). In addition, an analytical approach to quantitively assess the periodicity of individual components and their position relative to actin rings is also described.

Tropomyosin Tpm3.1 Is Required to Maintain the Structure and Function of the Axon Initial Segment.

The axon initial segment (AIS) is the site of action potential initiation and serves as a cargo transport filter and diffusion barrier that helps maintain neuronal polarity. The AIS actin cytoskeleton comprises actin patches and periodic sub-membranous actin rings. We demonstrate that tropomyosin isoform Tpm3.1 co-localizes with actin patches and that the inhibition of Tpm3.1 led to a reduction in the density of actin patches. Furthermore, Tpm3.1 showed a periodic distribution similar to sub-membranous actin rings but Tpm3.1 was only partially congruent with sub-membranous actin rings. Nevertheless, the inhibition of Tpm3.1 affected the uniformity of the periodicity of actin rings. Furthermore, Tpm3.1 inhibition led to reduced accumulation of AIS structural and functional proteins, disruption in sorting somatodendritic and axonal proteins, and a reduction in firing frequency. These results show that Tpm3.1 is necessary for the structural and functional maintenance of the AIS.

MIM-Deficient Mice Exhibit Anatomical Changes in Dendritic Spines, Cortex Volume and Brain Ventricles, and Functional Changes in Motor Coordination and Learning.

In this study, we performed a comprehensive behavioral and anatomical analysis of the Missing in Metastasis (Mtss1/MIM) knockout (KO) mouse brain. We also analyzed the expression of MIM in different brain regions at different ages. MIM is an I-BAR containing membrane curving protein, shown to be involved in dendritic spine initiation and dendritic branching in Purkinje cells in the cerebellum. Behavioral analysis of MIM KO mice revealed defects in both learning and reverse-learning, alterations in anxiety levels and reduced dominant behavior, and confirmed the previously described deficiency in motor coordination and pre-pulse inhibition. Anatomically, we observed enlarged brain ventricles and decreased cortical volume. Although MIM expression was relatively low in hippocampus after early development, hippocampal pyramidal neurons exhibited reduced density of thin and stubby dendritic spines. Learning deficiencies can be connected to all detected anatomical changes. Both behavioral and anatomical findings are typical for schizophrenia mouse models.

How Patients Talk About Care? Identifying Patient Experience Expressions from Online Discussions.

Patient experience is an emerging concept that supports the improvement of healthcare services through identified patient expectations and experiences. In addition to structured feedback through official channels, experiences about healthcare appear increasingly in digital services and social media. We explore a new patient experience harvesting process based on linguistic patterns to identify relevant expressions in online discussions about children's health. Our results from the analysis of 98 229 unique sentences suggests that the 7-step process can be useful in discovering patients' evaluations of their care experiences. We propose ways to extend the process to other care contexts by adjusting the semantic reference models.