Sex-related patterns in the electroencephalogram and their relevance in machine learning classifiers.

Deep learning is increasingly being proposed for detecting neurological and psychiatric diseases from electroencephalogram (EEG) data but the method is prone to inadvertently incorporate biases from training data and exploit illegitimate patterns. The recent demonstration that deep learning can detect the sex from EEG implies potential sex-related biases in deep learning-based disease detectors for the many diseases with unequal prevalence between males and females. In this work, we present the male- and female-typical patterns used by a convolutional neural network that detects the sex from clinical EEG (81% accuracy in a separate test set with 142 patients). We considered neural sources, anatomical differences, and non-neural artifacts as sources of differences in the EEG curves. Using EEGs from 1140 patients, we found electrocardiac artifacts to be leaking into the supposedly brain activity-based classifiers. Nevertheless, the sex remained detectable after rejecting heart-related and other artifacts. In the cleaned data, EEG topographies were critical to detect the sex, but waveforms and frequencies were not. None of the traditional frequency bands was particularly important for sex detection. We were able to determine the sex even from EEGs with shuffled time points and therewith completely destroyed waveforms. Researchers should consider neural and non-neural sources as potential origins of sex differences in their data, they should maintain best practices of artifact rejection, even when datasets are large, and they should test their classifiers for sex biases.

Further Dimensions for Sensing in Biofluids: Distinguishing Bioorganic Analytes by the Salt-Induced Adaptation of a Cucurbit[7]uril-Based Chemosensor.

Insufficient binding selectivity of chemosensors often renders biorelevant metabolites indistinguishable by the widely used indicator displacement assay. Array-based chemosensing methods are a common workaround but require additional effort for synthesizing a chemosensor library and setting up a sensing array. Moreover, it can be very challenging to tune the inherent binding preference of macrocyclic systems such as cucurbit[n]urils (CBn) by synthetic means. Using a novel cucurbit[7]uril-dye conjugate that undergoes salt-induced adaptation, we now succeeded in distinguishing 14 bioorganic analytes from each other through the facile stepwise addition of salts. The salt-specific concentration-resolved emission provides additional information about the system at a low synthetic effort. We present a data-driven approach to translate the human-visible curve differences into intuitive pairwise difference measures. Ion mobility experiments combined with density functional theory calculations gave further insights into the binding mechanism and uncovered an unprecedented ternary complex geometry for CB7. TThis work introduces the non-selectively binding, salt-adaptive cucurbit[n]uril system for sensing applications in biofluids such as urine, saliva, and blood serum.

Experiences from treating seven adult 5q spinal muscular atrophy patients with Nusinersen.

BACKGROUND: The antisense oligonucleotide Nusinersen recently became the first approved drug against spinal muscular atrophy (SMA). It was approved for all ages, albeit the clinical trials were conducted exclusively on children. Hence, clinical data on adults being treated with Nusinersen is scarce. In this case series, we report on drug application, organizational demands, and preliminary effects during the first 10 months of treatment with Nusinersen in seven adult patients. METHODS: All patients received intrathecal injections with Nusinersen. In cases with severe spinal deformities, we performed computed tomography (CT)-guided applications. We conducted a total of 40 administrations of Nusinersen. We evaluated the patients with motor, pulmonary, and laboratory assessments, and tracked patient-reported outcome. RESULTS: Intrathecal administration of Nusinersen was successful in most patients, even though access to the lumbar intrathecal space in adults with SMA is often challenging. No severe adverse events occurred. Six of the seven patients reported stabilization of motor function or reduction in symptom severity. The changes in the assessed scores did not reach a significant level within this short time period. CONCLUSIONS: Treating adult SMA patients with Nusinersen is feasible and most patients consider it beneficial. It demands a complex organizational and interdisciplinary effort. Due to the slowly decreasing motor functions in adult SMA patients, long observation phases for this recently approved treatment are needed to allow conclusions about effectiveness of Nusinersen in adults.

Influence of tissue conductivity changes on the EEG signal in the human brain: a simulation study.

Tissue disorders due to brain pathologies, like tumors, ischemia, or vasogenic edema, are known to impact the propagation of electrical fields. By using the finite element method the EEG forward problem was solved within an adapted subspace of a simplified human head model. Simulated electric potentials on the scalp revealed strong influences on the magnitude of the signal in almost all cases, even for ischemic tissue in which conductivity is lower than in healthy tissue. Remarkably, due to the "shunting effect" and the diminishing anisotropy of tissue conductivity, the signal amplitude of a radial dipole located in a sulcus was found to be higher than the signal of a dipolar source on a gyrus if the ischemic area was located underneath. The results demonstrate that pathological tissue changes have to be taken into account when evaluating EEG signals, especially when performing source localization.