Mechanical sensors, often backed by computer integration of input, apparently exist to detect whatever humans can sense, and beyond. We use such machines, from binoculars to hearing aids to sonar and more, to aid us in detecting things both within and beyond our normal range of detection.

Do you think human beings perceive anything that cannot be detected and measured, other than whatever subjective meaning we attach to what we perceive [so far]?

Probably yes. I'd say that our knowledge of ourselves is shrinking as our knowledge of the world's increasing complexity grows. Modern people often wouldn't listen to their own body signals unless they are confirmed by some sort of detectors. Hence there are plenty of such things as dowsing rods, gemstone pendulums, and other nontraditional "detectors", which are trying to fill the gaps of modern science methods.

Many electronic sensors will detect physical phenomenon with greater speed, dynamic range and areas of the electromagnetic and audio spectra, which cannot be detected by human senses. The data can also be calibrated against known references for accurate scientifically measurable values.

The main problem is that these are just data sets not information or intelligence data sets on which to base a decision making process. Artificial intelligence algorithms are still quite laughably immature. An example would be depth perception and edge detection. Humans process millions of visual data sets with ease everyday and although there have been improvements in such devices such as IR guidance systems for anti-aircraft missiles and X-box games controllers, they cannot compete with the wide aspect visual and speed capabilities of the human visual cortex.

Again music and speech audio are dealt with easily providing meaning to the human listener, but to an electronic system is just meaningless white noise.

Having worked and developed algorithms for Dynamic thermographic, laser Doppler flowmetry and photoplethysmography imaging for medical devices, the difficulty was always the interpretation of the data sets to pull out the information relevant for the decision maker. Inevitably the use of statistical methods such as non linear regression analysis was required. i.e. a best guess.

The important part of the original discussion topic was:

The nonlinear regression analysis doesn't fit

No...because of your word "perceive".

Another question, are humans beings "affected" by anything that cannot be detected and measured? Yes, I definitely think there are things that we cannot detect or measure, but that still affect us in major ways.

I'm not just talking about things like earthquakes and unpredictable weather. I'm talking more along the lines of chaos theory generally. For example, if the parents of an incredible genius are mean to their child, how will the world be affected later, as opposed to if this incredible genius grew up in a loving household? Can we predict if the kid will grow up to be a Bill Gates or instead a Bernie Madoff? Can the affect of this child's upbringing on the world's economy be measured one way or the other? I don't think so...chaos!!!

Machines can detect everything a human can, and much of what a human can't. But you can't confuse detection and sensing with vision, the ability to produce a model of the physical reality, or understanding complicated stuff like knowing that a wooden block set on top of a table is not part of the table. Or that a red circle drawn on a wall is not the same thing as a red rubber ball. Stuff a two year old handles reliably.

No. I'm not religious or mystical and I don't think humans have any magic non-physical senses. I don't believe there are things like vampires that can seen but not recorded on film, for example. No ghosts or souls.

What we do have is pattern recognition, and we are often better at knowing what to pay attention to. We can be better at understanding the significance of what we're sensing.