Portal field news

Portal field news

in ,

📢 | Dust-proof and drip-proof portable radio speaker Tivoli Audio "PAL BT" New barge ...


Dust-proof and drip-proof portable radio speaker Tivoli Audio "PAL BT" New barge ...

If you write the contents roughly
It features "good reception sensitivity" and "easy-to-use design" that emphasizes operability.

NAVYS Inc. (Location: Minato-ku, Tokyo, Representative Director: Noboru Takasaki) has recently upgraded "Tivo ... → Continue reading


Atpress is a pioneer in the distribution of press releases.
We handle a wide range of genres of news, from tourism and gourmet to entertainment and gadgets.
It is packed with the latest information that can be delivered faster than anywhere else because @Press, which distributes press releases to TV, magazines, web media, etc. and is strong in SNS!

Wikipedia related words

If there is no explanation, there is no corresponding item on Wikipedia.

感 度

感 度(Sensitivity,British: sensitivity) Is one of the statistical concepts.Although the definition differs depending on the field, it is an index related to "the relationship between the stimulus given to a certain object and the response to it".

Sensitivity of measuring equipment

Measuring equipmentIn and in the field of use, the word "sensitivity" has two main meanings:[1].

Sensitivity limit (sensitivity limit)

It means the minimum amount that a measuring device can measure.resolutionIt is sometimes called.In other words, the difference between the measured values ​​below this value is the amount that is considered meaningless.What sets the sensitivity limit is a very difficult issue, and there are some standards for industrial products.International Standards OrganizationAlthough it may be stipulated by such factors, it may become a major academic problem in itself, as there are various theories depending on the measuring instrument and the measurement target.Normally,off the shelfThe value is described in the measuring device of.

To give a familiar example, when it comes to the question of which is better, a person with a score of 1 or a person with a score of 100 in a 50-point test in 51-point increments (there may be a boundary between passing and failing). (Except for the case) Many people think that it is nonsense.It is a kind of story that is commonly called "comparing the backs of acorns".The reason why the difference between the two is considered meaningless is whether the order is maintained when the next similar test is performed (assuming that both did not do any act that makes a difference in study etc. during that time). Because there is a question.Differences in values ​​indicated by measuring instruments may also be nonsense differences in the same sense.For example, when the length of an object (let's call it a stick here) is measured by a measuring rod that measures the length by some principle.[2], The first measurement is 1 [cm] and the second is 31.52 [cm].In this case, it is reasonable to think that a value with at least two decimal places does not make sense from "one measurement value" alone.By the same reason, it is meaningless to discuss which of the object A with the measured value of 2 [cm] and the object with the measured value of 31.51 [cm] is longer from the single measured value of this measuring rod.Simple, measure the same object a sufficient number of times.[3],Thatstandard deviationIs defined as the "sensitivity limit".However, in the method where the sensitivity limit is set by the standard deviation (σ) or how many times it (2σ, 3σ in some cases, 6σ), "Accidental errorOther thanerrorDoes not exist. "

Even a difference that is meaningless from "one measurement" may make sense when "measured 1 times".Generally measured N times andaverageThere is a method of using the above as the measured value.In that case, how much differenceSignificantIs it usuallyStandard errorMay be used.In this case, the average value of the measured values ​​may be an infinite decimal such as 32.5172… cm, but if the standard error is 0.01 [cm],Significant figuresIs a two decimal place, and rounded down to 2 ± 31.52 cm.

In addition,Frequency characteristicIn some cases, it may be considered in consideration of such factors.For example, noise of 500 Hz or higher is always recorded at about 200 mVpp, but it is quite possible that a difference of about 1 mVpp is reliable as a single measurement value of the DC component.

Sensitivity coefficient (sensitivity coefficient)

Measuring equipmentThe ratio of the size of the device to the output value of the measuring device.Of the oscilloscoperangeIs this.Usually variable.

Detection limit(Synonymous with sensitivity limit in analytical chemistry)

Assuming that the measured values ​​are not correlated with each other and have a normal distribution, the limit is 3 times the standard deviation of the blank measured value (Kaiser's limit) or 3.29 times (Currie's limit). ..

Relationship between device noise and detection limit

The cause of the fluctuation of the measured value is the noise of the measuring equipment as well as the slight fluctuation of the experimental conditions.In continuous measurement, it appears as line fluctuation.Even if the signal is amplified, the noise is also amplified, so that it is difficult to understand if the signal is not sufficiently larger than the noise, and the noise becomes a factor that limits the sensitivity.

The definition of equipment noise is as follows: (1) rms noise The square root of the sum of squares of fluctuations from the average value divided by n-2 (1) peak to peak noise Difference between the maximum and minimum values ​​within a certain period of time , Or there is an average of the difference between the maximum and minimum values ​​of multiple intervals, and (2) is used theoretically, but it is usually indicated by (1). (2) is the same formula as the standard deviation, but in rms noise where the time between observation points is extremely short, it is between the measured values.AutocorrelationTherefore, it cannot be treated the same as the standard deviation to give a mathematically exact detection limit.However, if there is no autocorrelation, (2) is said to be about 1 times that of (5), so (2) seems to indicate a noise range, and if it is 2 to 3 times that distance, there is a significant difference and there is almost no error. it is conceivable that. (""noiseRefer to "Measuring instrument noise" in the "Measuring equipment noise" section.

Sensitivity in medicine

In medicine感 度IsClinical examinationIt is one of the indexes that determine the character of a person, and is a value defined as "the probability that something that should be judged as positive is correctly judged as positive" for a certain test.

High sensitivity (high sensitivity) means that "what should be judged as positive is likely to be correctly judged as positive" or "what should be judged as positive is likely to be judged as negative by mistake". It means "low sex".

Sensitivity and specificity

For expressions that are paired with sensitivitySpecificityAnd specificity is related to sensitivity.A specific example will be described below.

With a certain illness AserumThe value inside rises酵素Suppose.In this test, in normal peopleaverageIt is a number of about 100, but for those who have illness A, it will rise significantly to a number of about 1000 on average.statisticsLet's say you know it.In this case, where is the value, that is, the boundary between normal and abnormal?

For example, assume that 150 or more is abnormal and less than 150 is normal, and this inspection is operated.Then, even though I wasn't really ill A,AnomalyThe number of subjects judged to be "" will inevitably increase (false positiveWill increase).Such tests are unlikely to miss people who have illness A, but are unlikely to correctly determine who does not have illness A.That is, it is a high-sensitivity and low-specificity test.

Even with the exact same test, 800 or more is abnormal, and less than 800 is normal.正常The number of subjects judged to be "" will increase (False negativeWill increase).Such a test is a low-sensitivity, high-specificity test that is unlikely to unnecessarily worry people who do not have disease A, but cannot correctly determine who has disease A.

An ideal test is one that is completely 100% sensitive and specific, but in reality there is no such perfect test.As a general rule, the cutoff value should be an appropriate value so that both the sensitivity and specificity values ​​are as high as possible.However, the reality is that it is adjusted according to the purpose of the inspection.For example JapanMad cow diseaseSo, first of all,ELISABy lawScreening testThis is an inexpensive test that greatly increases sensitivity, reduces the possibility of overlooking positives as much as possible, and sacrifices specificity (that is, false positives are likely to occur).


For sensitivity, specificity, positive predictive value, and negative predictive value, refer to the table below.

True state
(Determined by the result of detailed examination such as biopsy)
PositiveTrue positivefalse positive
(Type I error: α error)
Positive predictive value =
Number of true positives
   Number of positive tests
negativeFalse negative
(Type II error: β error)
True negativeNegative predictive value =
Number of true negatives
 Number of negative tests
Sensitivity =
Number of true positives
Total of really positive people
Specificity =
Number of true negatives
Total of really negative people

Related item


  • Japan Epidemiology Society (2000): Epidemiology Dictionary 3rd Edition, Japan Public Health Association
  • MH Katz (2006): Study Design and Statistical Analysis: A Practical Guide for Clinicians, Cambridge University Press, 2006. (Japanese translation) Statistical Practice Guide for Clinical Research-From Paper Planning to Submission EDIXi Publishing Department 2011

Photo sensitivity (ISO speed)

What is ISO sensitivity (ISO speed, ISO speed)?the filmOne of the international standards of. It is represented by numbers such as ISO100, ISO400, and ISO800. The larger the number, the worse the image quality will be, although you can shoot even in dark places.Digital cameraThen, it may be used in the same way as "equivalent to ISO100".

Wireless sensitivity

Antenna sensitivity

antennaPerformance (gain), But it is not the correct way to say it.

Receiver sensitivity

In an environment where the external noise level is sufficiently low, the better the sensitivity, the weaker the radio wave can be received.Standard signal generatorMeasure using.

It is defined as the minimum input signal for ensuring the quality of the received signal required for the wireless system, and specifically, it is measured as a pair of the quality evaluation index such as SINAD and BER and the input signal strength.

  • Narrow FM receiver
    • SINAD (SIgnal to Noise And Distortion) Sensitivity
    • Noise suppression sensitivity
    • Squelch感 度
  • Digital receiver
    • BER(Bit error rate)
    • FER (frame error rate)
    • PER (Packet Error Rate)

Customary expression during wireless communication

It is used to roughly report the reception status to the partner station at the start of wireless communication.

  • Good sensitivity.
  • Both impressions are good. (Abbreviation for good sensitivity and intelligibility)

Sensitivity in explosives

Sensitivity is a measure of explosive sensitivity to external stimuli of explosives.If an explosive explodes easily in response to an external stimulus, the explosive is said to be sensitive to that stimulus.On the contrary, if it does not explode easily, it is expressed as insensitive or insensitive.Stimuli include shock, friction, heat, flames, and sparks.Sensitivity also changes due to external factors such as temperature, humidity, and atmospheric pressure.It also changes due to air bubbles and impurities inside the explosive.Since the sensitivity of explosives is important for security, various sensitivity tests are performed and measured strictly.


[How to use footnotes]
  1. ^ Masaaki Ogasawara "Basics of Measurement and Data Analysis" Tokyo Kagaku Dojin
  2. ^ For example, in the case of an interferometer that measures the length using the interference of waves (sound waves, etc.), the primary "length" is replaced with the secondary "interference intensity", and usually it is replaced with ( With a microphone etc.)VoltageSince it is converted to and read by a voltmeter, the sensitivity limit of the voltmeter in the subsequent stage and the sensitivity limit of the entire interferometer are worse than the sensitivity limit of the interferometer.However, since the display is popped out based on the sensitivity limit of the voltmeter, it is necessary to clarify from various measurements how much the number is meaningful.Such considerations are generally very difficult and in some cases哲学It may be a problem.
  3. ^ When a sufficient number of measurements are taken, the standard deviation of the measured values ​​converges to a constant value.


Back to Top