Investigating Explanatory Factors of Machine Learning Models for Plant Classification.

Recent progress in machine learning and deep learning has enabled the implementation of plant and crop detection using systematic inspection of the leaf shapes and other morphological characters for identification systems for precision farming. However, the models used for this approach tend to become black-box models, in the sense that it is difficult to trace characters that are the base for the classification. The interpretability is therefore limited and the explanatory factors may not be based on reasonable visible characters. We investigate the explanatory factors of recent machine learning and deep learning models for plant classification tasks. Based on a Daucus carota and a Beta vulgaris image data set, we implement plant classification models and compare those models by their predictive performance as well as explainability. For comparison we implemented a feed forward convolutional neuronal network as a default model. To evaluate the performance, we trained an unsupervised Bayesian Gaussian process latent variable model as well as a convolutional autoencoder for feature extraction and rely on a support vector machine for classification. The explanatory factors of all models were extracted and analyzed. The experiments show, that feed forward convolutional neuronal networks (98.24% and 96.10% mean accuracy) outperforms the Bayesian Gaussian process latent variable pipeline (92.08% and 94.31% mean accuracy) as well as the convolutional autoenceoder pipeline (92.38% and 93.28% mean accuracy) based approaches in terms of classification accuracy, even though not significant for Beta vulgaris images. Additionally, we found that the neuronal network used biological uninterpretable image regions for the plant classification task. In contrast to that, the unsupervised learning models rely on explainable visual characters. We conclude that supervised convolutional neuronal networks must be used carefully to ensure biological interpretability. We recommend unsupervised machine learning, careful feature investigation, and statistical feature analysis for biological applications.

Identifying geographically differentiated features of Ethopian Nile tilapia (Oreochromis niloticus) morphology with machine learning.

Visual characteristics are among the most important features for characterizing the phenotype of biological organisms. Color and geometric properties define population phenotype and allow assessing diversity and adaptation to environmental conditions. To analyze geometric properties classical morphometrics relies on biologically relevant landmarks which are manually assigned to digital images. Assigning landmarks is tedious and error prone. Predefined landmarks may in addition miss out on information which is not obvious to the human eye. The machine learning (ML) community has recently proposed new data analysis methods which by uncovering subtle features in images obtain excellent predictive accuracy. Scientific credibility demands however that results are interpretable and hence to mitigate the black-box nature of ML methods. To overcome the black-box nature of ML we apply complementary methods and investigate internal representations with saliency maps to reliably identify location specific characteristics in images of Nile tilapia populations. Analyzing fish images which were sampled from six Ethiopian lakes reveals that deep learning improves on a conventional morphometric analysis in predictive performance. A critical assessment of established saliency maps with a novel significance test reveals however that the improvement is aided by artifacts which have no biological interpretation. More interpretable results are obtained by a Bayesian approach which allows us to identify genuine Nile tilapia body features which differ in dependence of the animals habitat. We find that automatically inferred Nile tilapia body features corroborate and expand the results of a landmark based analysis that the anterior dorsum, the fish belly, the posterior dorsal region and the caudal fin show signs of adaptation to the fish habitat. We may thus conclude that Nile tilapia show habitat specific morphotypes and that a ML analysis allows inferring novel biological knowledge in a reproducible manner.

Effects of negative air ions on oxygen uptake kinetics, recovery and performance in exercise: a randomized, double-blinded study.

Limited research has suggested that acute exposure to negatively charged ions may enhance cardio-respiratory function, aerobic metabolism and recovery following exercise. To test the physiological effects of negatively charged air ions, 14 trained males (age: 32 ± 7 years; VO2max: 57 ± 7 mL min(-1) kg(-1)) were exposed for 20 min to either a high-concentration of air ions (ION: 220 ± 30 × 10(3) ions cm(-3)) or normal room conditions (PLA: 0.1 ± 0.06 × 10(3) ions cm(-3)) in an ionization chamber in a double-blinded, randomized order, prior to performing: (1) a bout of severe-intensity cycling exercise for determining the time constant of the phase II VO2 response (τ) and the magnitude of the VO2 slow component (SC); and (2) a 30-s Wingate test that was preceded by three 30-s Wingate tests to measure plasma [adrenaline] (ADR), [nor-adrenaline] (N-ADR) and blood [lactate] (B(Lac)) over 20 min during recovery in the ionization chamber. There was no difference between ION and PLA for the phase II VO2 τ (32 ± 14 s vs. 32 ± 14 s; P = 0.7) or VO2 SC (404 ± 214 mL vs 482 ± 217 mL; P = 0.17). No differences between ION and PLA were observed at any time-point for ADR, N-ADR and B(Lac) as well as on peak and mean power output during the Wingate tests (all P > 0.05). A high-concentration of negatively charged air ions had no effect on aerobic metabolism during severe-intensity exercise or on performance or the recovery of the adrenergic and metabolic responses after repeated-sprint exercise in trained athletes.

Cardiovascular oscillations of the carotid artery assessed by magnetoelastic skin curvature sensor.

The present study concerns the nondisturbing assessment of cardiovascular oscillations of the carotid artery using a novel skin curvature sensor on the neck. The mechanical oscillations of the skin reflect changes of the artery radius and thus relevant physiological data such as cardiac and respiratory activities, their mutual dependencies, and even changes of blood pressure. The skin curvature sensor is easy to handle and it minimally disturbs the patient, which is relevant for many medical areas such as sleep monitoring.