A large number of papers are appearing in the biomedical engineering literature that describe the use of machine learning techniques to develop classifiers for detection or diagnosis of disease. However, the usefulness of this approach in developing clinically validated diagnostic techniques so far has been limited and the methods are prone to overfitting and other problems which may not be immediately apparent to the investigators. This commentary is intended to help sensitize investigators as well as readers and reviewers of papers to some potential pitfalls in the development of classifiers, and suggests steps that researchers can take to help avoid these problems. Building classifiers should be viewed not simply as an add-on statistical analysis, but as part and parcel of the experimental process. Validation of classifiers for diagnostic applications should be considered as part of a much larger process of establishing the clinical validity of the diagnostic technique.

The Precautionary Principle has become enshrined in international law, and is the basis for European environmental legislation. However, "precautionary" decisions have been controversial, and the principle itself lacks clear definition. A recent commentary by the European Commission offers guidelines for politically transparent application of the principle, while emphasizing the need for careful review of relevant scientific data. Recent precautionary policies for limiting public exposure to radio-frequency energy shows that the principle can be applied in a way that does not conflict with traditional exposure guidelines. Major uncertainties still remain in the standard of proof needed to invoke the principle.

This article reviews thermal mechanisms of interaction between radiofrequency (RF) fields and biological systems, focusing on theoretical frameworks that are of potential use in setting guidelines for human exposure to RF energy. Several classes of thermal mechanisms are reviewed that depend on the temperature increase or rate of temperature increase and the relevant dosimetric considerations associated with these mechanisms. In addition, attention is drawn to possible molecular and physiological reactions that could be induced by temperature elevations below 0.1 degrees, which are normal physiological responses to heat, and to the so-called microwave auditory effect, which is a physiologically trivial effect resulting from thermally-induced acoustic stimuli. It is suggested that some reported "nonthermal" effects of RF energy may be thermal in nature; also that subtle thermal effects from RF energy exist but have no consequence to health or safety. It is proposed that future revisions of exposure guidelines make more explicit use of thermal models and empirical data on thermal effects in quantifying potential hazards of RF fields.

This survey measured radiofrequency (RF) fields from wireless local area networks (WLANs) using Wi-Fi technology against a background of RF fields in the environment over the frequency range 75 MHz-3 GHz. A total of 356 measurements were conducted at 55 sites (including private residences, commercial spaces, health care and educational institutions, and other public spaces) in four countries (U.S., France, Germany, Sweden). Measurements were conducted under conditions that would result in the higher end of exposures from such systems. Where possible, measurements were conducted in public spaces as close as practical to the Wi-Fi access points. Additional measurements were conducted at a distance of approximately 1 m from a laptop while it was uploading and downloading large files to the WLAN. This distance was chosen to allow a useful comparison of fields in the far-field of the antenna in the laptop, and give a representative measure of the exposure that a bystander might receive from the laptop. The exposure to the user, particularly if the antenna of the client card were placed against his or her body, would require different measurement techniques beyond the scope of this study. In all cases, the measured Wi-Fi signal levels were very far below international exposure limits (IEEE C95.1-2005 and ICNIRP) and in nearly all cases far below other RF signals in the same environments. An discusses technical aspects of the IEEE 802.11 standard on which WLANs operate that are relevant to determining the levels of RF energy exposure from WLANs. Important limiting factors are the low operating power of client cards and access points, and the low duty cycle of transmission that normally characterizes their operation.

In September 2017, the U.S. Food and Drug Administration (FDA) made a striking announcement. Transforming its current regulatory practice for approving and certifying medical devices-the FDA announced a bold new plan, the Digital Health Software Precertification (Precert) Program, to offer an entirely new regulatory model to assess smartphone apps, wearables, sensors, and software. This transformation and medicalization of the consumer health market present both opportunities and obstacles, by opening up large markets for health monitoring and diagnosis using inexpensive mass-market, off-the-shelf devices. It also raises challenges, both related to privacy and effective uses of the devices to promote health. The Fitbit and Apple Watch are examples.