Discovery of different metabotypes in overconditioned dairy cows by means of machine learning.

Using data from targeted metabolomics in serum in combination with machine learning (ML) approaches, we aimed at (1) identifying divergent metabotypes in overconditioned cows and at (2) exploring how metabotypes are associated with lactation performance, blood metabolites, and hormones. In a previously established animal model, 38 pregnant multiparous Holstein cows were assigned to 2 groups that were fed differently to reach either high (HBCS) or normal (NBCS) body condition score (BCS) and backfat thickness (BFT) until dryoff at -49 d before calving [NBCS: BCS < 3.5 (3.02 ± 0.24) and BFT < 1.2 cm (0.92 ± 0.21), mean ± SD; HBCS: BCS > 3.75 (3.82 ± 0.33) and BFT > 1.4 cm (2.36 ± 0.35)]. Cows were then fed the same diets during the dry period and the subsequent lactation, and maintained the differences in BFT and BCS throughout the study. Blood samples were collected weekly from 7 wk antepartum (ap) to 12 wk postpartum (pp) to assess serum concentrations of metabolites (by targeted metabolomics and by classical analyses) and metabolic hormones. Metabolic clustering by applying 4 supervised ML-based classifiers [sequential minimal optimization (SMO), random forest (RF), alternating decision tree (ADTree), and naïve Bayes-updatable (NB)] on the changes (d 21 pp minus d 49 ap) in concentrations of 170 serum metabolites resulted in 4 distinct metabolic clusters: HBCS predicted HBCS (HBCS-PH, n = 13), HBCS predicted NBCS (HBCS-PN, n = 6), NBCS predicted NBCS (NBCS-PN, n = 15), and NBCS predicted HBCS (NBCS-PH, n = 4). The accuracies of SMO, RF, ADTree, and NB classifiers were >70%. Because the number of NBCS-PH cows was low, we did not consider this group for further comparisons. Dry matter intake (kg/d and percentage of body weight) and energy intake were greater in HBCS-PN than in HBCS-PH in early lactation, and HBCS-PN also reached a positive energy balance earlier than did HBCS-PH. Milk yield was not different between groups, but milk protein percentage was greater in HBCS-PN than in HBCS-PH cows. The circulating concentrations of fatty acids (FA) increased during early lactation in both groups, but HBCS-PN cows had lower concentrations of ß-hydroxybutyrate, indicating lower ketogenesis compared with HBCS-PH cows. The concentrations of insulin, insulin-like growth factor 1, leptin, adiponectin, haptoglobin, glucose, and revised quantitative insulin sensitivity check index did not differ between the groups, whereas serum concentrations of glycerophospholipids were lower before calving in HBCS-PH than in HBCS-PN cows. Glycine was the only amino acid that had higher concentration after calving in HBCS-PH than in HBCS-PN cows. The circulating concentrations of some short- (C2, C3, and C4) and long-chain (C12, C16:0, C18:0, and C18:1) acylcarnitines on d 21 pp were greater in HBCS-PH than in HBCS-PN cows, indicating incomplete FA oxidation. In conclusion, the use of ML approaches involving data from targeted metabolomics in serum is a promising method for differentiating divergent metabotypes from apparently similar BCS phenotypes. Further investigations, using larger numbers of cows and farms, are warranted for confirmation of this finding.

Metabolomics meets machine learning: Longitudinal metabolite profiling in serum of normal versus overconditioned cows and pathway analysis.

This study aimed to investigate the differences in the metabolic profiles in serum of dairy cows that were normal or overconditioned when dried off for elucidating the pathophysiological reasons for the increased health disturbances commonly associated with overconditioning. Fifteen weeks antepartum, 38 multiparous Holstein cows were allocated to either a high body condition (HBCS; n = 19) group or a normal body condition (NBCS; n = 19) group and were fed different diets until dry-off to amplify the difference. The groups were also stratified for comparable milk yields (NBCS: 10,361 ± 302 kg; HBCS: 10,315 ± 437 kg; mean ± standard deviation). At dry-off, the cows in the NBCS group (parity: 2.42 ± 1.84; body weight: 665 ± 64 kg) had a body condition score (BCS) <3.5 and backfat thickness (BFT) <1.2 cm, whereas the HBCS cows (parity: 3.37 ± 1.67; body weight: 720 ± 57 kg) had BCS >3.75 and BFT >1.4 cm. During the dry period and the subsequent lactation, both groups were fed identical diets but maintained the BCS and BFT differences. A targeted metabolomics (AbsoluteIDQ p180 kit, Biocrates Life Sciences AG, Innsbruck, Austria) approach was performed in serum samples collected on d -49, +3, +21, and +84 relative to calving for identifying and quantifying up to 188 metabolites from 6 different compound classes (acylcarnitines, AA, biogenic amines, glycerophospholipids, sphingolipids, and hexoses). The concentrations of 170 metabolites were above the limit of detection and could thus be used in this study. We used various machine learning (ML) algorithms (e.g., sequential minimal optimization, random forest, alternating decision tree, and naïve Bayes-updatable) to analyze the metabolome data sets. The performance of each algorithm was evaluated by a leave-one-out cross-validation method. The accuracy of classification by the ML algorithms was lowest on d 3 compared with the other time points. Various ML methods (partial least squares discriminant analysis, random forest, information gain ranking) were then performed to identify those metabolites that were contributing most significantly to discriminating the groups. On d 21 after parturition, 12 metabolites (acetylcarnitine, hexadecanoyl-carnitine, hydroxyhexadecenoyl-carnitine, octadecanoyl-carnitine, octadecenoyl-carnitine, hydroxybutyryl-carnitine, glycine, leucine, phosphatidylcholine-diacyl-C40:3, trans-4-hydroxyproline, carnosine, and creatinine) were identified in this way. Pathway enrichment analysis showed that branched-chain AA degradation (before calving) and mitochondrial ß-oxidation of long-chain fatty acids along with fatty acid metabolism, purine metabolism, and alanine metabolism (after calving) were significantly enriched in HBCS compared with NBCS cows. Our results deepen the insights into the phenotype related to overconditioning from the preceding lactation and the pathophysiological sequelae such as increased lipolysis and ketogenesis and decreased feed intake.

Prediction of memory formation based on absolute electroencephalographic phases in rhinal cortex and hippocampus outperforms prediction based on stimulus-related phase shifts.

Absolute (i.e. measured) rhinal and hippocampal phase values are predictive for memory formation. It has been an open question, whether the capability of mediotemporal structures to react to stimulus presentation with phase shifts may be similarly indicative of successful memory formation. We analysed data from 27 epilepsy patients implanted with depth electrodes in the hippocampus and entorhinal cortex, who performed a continuous word recognition task. Electroencephalographic phase information related to the first presentation of repeatedly presented words was used for prediction of subsequent remembering vs. forgetting applying a support vector machine. The capability to predict successful memory formation based on stimulus-related phase shifts was compared to that based on absolute phase values. Average hippocampal phase shifts were larger and rhinal phase shifts were more accumulated for later remembered compared to forgotten trials. Nevertheless, prediction based on absolute phase values clearly outperformed phase shifts and there was no significant increase in prediction accuracies when combining both measures. Our findings indicate that absolute rhinal and hippocampal phases and not stimulus-related phase shifts are most relevant for successful memory formation. Absolute phases possibly affect memory formation via influencing neural membrane potentials and thereby controlling the timing of neural firing.

Prediction of successful memory encoding based on single-trial rhinal and hippocampal phase information.

Mediotemporal EEG characteristics are closely related to long-term memory formation. It has been reported that rhinal and hippocampal EEG measures reflecting the stability of phases across trials are better suited to distinguish subsequently remembered from forgotten trials than event-related potentials or amplitude-based measures. Theoretical models suggest that the phase of EEG oscillations reflects neural excitability and influences cellular plasticity. However, while previous studies have shown that the stability of phase values across trials is indeed a relevant predictor of subsequent memory performance, the effect of absolute single-trial phase values has been little explored. Here, we reanalyzed intracranial EEG recordings from the mediotemporal lobe of 27 epilepsy patients performing a continuous word recognition paradigm. Two-class classification using a support vector machine was performed to predict subsequently remembered vs. forgotten trials based on individually selected frequencies and time points. We demonstrate that it is possible to successfully predict single-trial memory formation in the majority of patients (23 out of 27) based on only three single-trial phase values given by a rhinal phase, a hippocampal phase, and a rhinal-hippocampal phase difference. Overall classification accuracy across all subjects was 69.2% choosing frequencies from the range between 0.5 and 50Hz and time points from the interval between -0.5s and 2s. For 19 patients, above chance prediction of subsequent memory was possible even when choosing only time points from the prestimulus interval (overall accuracy: 65.2%). Furthermore, prediction accuracies based on single-trial phase surpassed those based on single-trial power. Our results confirm the functional relevance of mediotemporal EEG phase for long-term memory operations and suggest that phase information may be utilized for memory enhancement applications based on deep brain stimulation.

Seizure-related modulation of systemic arterial blood pressure in focal epilepsy.

OBJECTIVE: Cardiorespiratory function alterations are commonly observed with epileptic seizures and may lead to syncope and sudden unexpected death in epilepsy (SUDEP). Although most previous research has focused on controlling heart rate (HR) and respiration, little is known about seizure-related regulation of systemic blood pressure (BP). Herein, we have investigated whether the periictal modulation of systemic BP and HR depends on seizure characteristics. METHODS: Systemic arterial BP, HR, and peripheral capillary oxygen saturation (SPO2 ) were continuously and noninvasively monitored using the ccNexfin device in those epilepsy patients undergoing video-electroencephalography (EEG) telemetry. Data are given as mean ± standard deviation (SD). RESULTS: Forty-five seizures in 37 patients were included. In focal seizures (FS, n = 35), the mean arterial BP (MAP) increased by 33 ± 35% and the HR by 53 ± 44%, whereas the SPO2 remained unaltered. The MAP and HR increases were significantly greater in FS with alterations in consciousness than in those without. For those FS that evolved to bilateral convulsive seizures (BCS, n = 10), all of the ictal recordings were compromised by artifacts. However, 2 min after seizure cessation, the MAP was enhanced by only 16 ± 14% and returned to a baseline slightly below preictal levels after 5 min, whereas the HR was increased by 77 ± 33% and remained elevated throughout the postictal phase. SIGNIFICANCE: Periictal regulation of systemic BP and HR displays distinct patterns depending on the type of seizure with focal onset. These changes were unrelated to alterations in SPO2 . The potential clinical implications of these findings are discussed in the article.

Performance of ECG-based seizure detection algorithms strongly depends on training and test conditions.

OBJECTIVE: To identify non-EEG-based signals and algorithms for detection of motor and non-motor seizures in people lying in bed during video-EEG (VEEG) monitoring and to test whether these algorithms work in freely moving people during mobile EEG recordings. METHODS: Data of three groups of adult people with epilepsy (PwE) were analyzed. Group 1 underwent VEEG with additional devices (accelerometry, ECG, electrodermal activity); group 2 underwent VEEG; and group 3 underwent mobile EEG recordings both including one-lead ECG. All seizure types were analyzed. Feature extraction and machine-learning techniques were applied to develop seizure detection algorithms. Performance was expressed as sensitivity, precision, F1 score, and false positives per 24 hours. RESULTS: The algorithms were developed in group 1 (35 PwE, 33 seizures) and achieved best results (F1 score 56%, sensitivity 67%, precision 45%, false positives 0.7/24 hours) when ECG features alone were used, with no improvement by including accelerometry and electrodermal activity. In group 2 (97 PwE, 255 seizures), this ECG-based algorithm largely achieved the same performance (F1 score 51%, sensitivity 39%, precision 73%, false positives 0.4/24 hours). In group 3 (30 PwE, 51 seizures), the same ECG-based algorithm failed to meet up with the performance in groups 1 and 2 (F1 score 27%, sensitivity 31%, precision 23%, false positives 1.2/24 hours). ECG-based algorithms were also separately trained on data of groups 2 and 3 and tested on the data of the other groups, yielding maximal F1 scores between 8% and 26%. SIGNIFICANCE: Our results suggest that algorithms based on ECG features alone can provide clinically meaningful performance for automatic detection of all seizure types. Our study also underscores that the circumstances under which such algorithms were developed, and the selection of the training and test data sets need to be considered and limit the application of such systems to unseen patient groups behaving in different conditions.

Combining SPECT and Quantitative EEG Analysis for the Automated Differential Diagnosis of Disorders with Amnestic Symptoms.

Single photon emission computed tomography (SPECT) and Electroencephalography (EEG) have become established tools in routine diagnostics of dementia. We aimed to increase the diagnostic power by combining quantitative markers from SPECT and EEG for differential diagnosis of disorders with amnestic symptoms. We hypothesize that the combination of SPECT with measures of interaction (connectivity) in the EEG yields higher diagnostic accuracy than the single modalities. We examined 39 patients with Alzheimer's dementia (AD), 69 patients with depressive cognitive impairment (DCI), 71 patients with amnestic mild cognitive impairment (aMCI), and 41 patients with amnestic subjective cognitive complaints (aSCC). We calculated 14 measures of interaction from a standard clinical EEG-recording and derived graph-theoretic network measures. From regional brain perfusion measured by 99mTc-hexamethyl-propylene-aminoxime (HMPAO)-SPECT in 46 regions, we calculated relative cerebral perfusion in these patients. Patient groups were classified pairwise with a linear support vector machine. Classification was conducted separately for each biomarker, and then again for each EEG- biomarker combined with SPECT. Combination of SPECT with EEG-biomarkers outperformed single use of SPECT or EEG when classifying aSCC vs. AD (90%), aMCI vs. AD (70%), and AD vs. DCI (100%), while a selection of EEG measures performed best when classifying aSCC vs. aMCI (82%) and aMCI vs. DCI (90%). Only the contrast between aSCC and DCI did not result in above-chance classification accuracy (60%). In general, accuracies were higher when measures of interaction (i.e., connectivity measures) were applied directly than when graph-theoretical measures were derived. We suggest that quantitative analysis of EEG and machine-learning techniques can support differentiating AD, aMCI, aSCC, and DCC, especially when being combined with imaging methods such as SPECT. Quantitative analysis of EEG connectivity could become an integral part for early differential diagnosis of cognitive impairment.