Signs may have no meaning to the patient, and may even go unnoticed, but may be meaningful and significant to the healthcare provider in assisting the diagnosis of medical condition(s) responsible for the patient's symptoms.
The art of interpreting clinical signs was originally called semiotics (a term now used for the study of sign communication in general) in English. This term, then written semeiotics (derived from the Greek adjective σημειοτικός: semeiotikos, "to do with signs"), was first used in English in 1670 by Henry Stubbes (1631–1676), to denote the branch of medical science relating to the interpretation of signs:
…nor is there any thing to be relied upon in Physick, but an exact knowledge of medicinal phisiology (founded on observation, not principles), semeiotics, method of curing, and tried (not excogitated, not commanding) medicines…
Signs are different from symptoms, the subjective experiences, such as fatigue, that patients might report to their examining physician.
For convenience, signs are commonly distinguished from symptoms as follows: Both are something abnormal, relevant to a potential medical condition, but a symptom is experienced and reported by the patient, while a sign is discovered by the physician during examination or by a clinical scientist by means of an in vitro examination of the patient.:75
A slightly different definition views signs as any indication of a medical condition that can be objectively observed (i.e., by someone other than the patient), whereas a symptom is merely any manifestation of a condition that is apparent to the patient (i.e., something consciously affecting the patient). From this definition, it can be said that an asymptomatic patient is uninhibited by disease. However, a doctor may discover the sign hypertension in an asymptomatic patient, who does not experience "dis-ease", and the sign indicates a disease state that poses a hazard to the patient. With this set of definitions, there is some overlap – certain things may qualify as both a sign and a symptom (e.g., a bloody nose).
Lester S. King, author of Medical Thinking, argues that an "essential feature" of a sign is that there is both a sign [or "signifier"] and a "thing signified". And, because "the essence of a sign is to convey information", it can only be a sign, properly speaking, if it has meaning. Therefore, "a sign ceases to be a sign when you cannot read it".:73–74 A person, who has and exercises the knowledge required to understand the significance or indication or meaning of the sign, is necessary for something to be a complete sign. A physical phenomenon that is not actually interpreted as a sign pointing to something else is, in medicine, merely a symptom. Thus, King rejects "these present-day views [distinguishing signs from symptoms based on patient-subjective versus clinician-objective], however widely accepted, as quite faulty, at variance not only with ordinary usage but with the entire history of medicine.":77
Medical signs may be classified by the type of inference that may be made from their presence,:80–81 for example:
Prognostic signs (from progignṓskein, προγιγνώσκειν, "to know beforehand"): signs that indicate the outcome of the current bodily state of the patient (i.e., rather than indicating the name of the disease). Prognostic signs always point to the future. Perhaps the most famous prognostic sign is the facies Hippocratica:
"[If the patient's facial] appearance may be described thus: the nose sharp, the eyes sunken, the temples fallen in, the ears cold and drawn in and their lobes distorted, the skin of the face hard, stretched and dry, and the colour of the face pale or dusky.… and if there is no improvement within [a prescribed period of time], it must be realized that this sign portends death."
Anamnestic signs (from anamnēstikós, ἀναμνηστικός, "able to recall to mind"): signs that (taking into account the current state of a patient's body), indicate the past existence of a certain disease or condition. Anamnestic signs always point to the past. (Whenever we see a man walking with a particular gait, with one arm paralysed in a particular way, we say "This man has had a stroke"; and, if we see a woman in her late 50s with one arm distorted in a particular way, we say "She had polio as a child".)[dubious– discuss]
Diagnostic signs (from diagnōstikós, διαγνωστικός, "able to distinguish"): signs that lead to the recognition and identification of a disease (i.e., they indicate the name of the disease).
Pathognomonic signs (from pathognomonikós, παθογνωμονικός, "skilled in diagnosis", derived from páthos, πάθος, "suffering, disease", and gnṓmon, γνώμον, "judge, indicator"): the particular signs whose presence means, beyond any doubt, that a particular disease is present. They represent a marked intensification of a diagnostic sign. (An example would be the palmar xanthomata seen on the hands of people suffering from hyperlipoproteinaemia.) Singular pathognomonic signs are relatively uncommon.
"[Thus] a symptom is a phenomenon, caused by an illness and observable directly in experience. We may speak of it as a manifestation of illness. When the observer reflects on that phenomenon and uses it as a base for further inferences, then that symptom is transformed into a sign. As a sign it points beyond itself — perhaps to the present illness, or to the past or to the future. That to which a sign points is part of its meaning, which may be rich and complex, or scanty, or any gradation in between. In medicine, then, a sign is thus a phenomenon from which we may get a message, a message that tells us something about the patient or the disease. A phenomenon or observation that does not convey a message is not a sign. The distinction between signs and symptom rests on the meaning, and this is not perceived but inferred.":81
Technological development creating signs detectable only by physicians
Prior to the nineteenth century there was little difference in the powers of observation between physician and patient. Most medical practice was conducted as a joint co-operative interaction between the physician and the patient as equals. Whilst each noticed much the same things, the physician had a more informed interpretation of those things: "the physicians knew what the findings meant and the layman did not".:82
However, the patient was gradually removed from the medical interaction due to significant technological advances such as:
The 1808 introduction of the percussion technique:
"The process through which "the physician can assess the state of the underlying lung by sensing the character of vibrations by gentle taps on the chest wall [something which] greatly facilitated the diagnosis of pneumonia and other respiratory diseases" 
The 1819 introduction by René Laënnec (1781–1826) of the technique of auscultation (using a stethoscope to listen to the circulatory and respiratory functions of the body). Laënnec's publication was translated into English, 1821–1834, by John Forbes.
The 1846 introduction by surgeon John Hutchinson (1811–1861) of the spirometer, an apparatus for assessing the mechanical properties of the lungs via measurements of forced exhalation and forced inhalation. (The recorded lung volumes and air flow rates are used to distinguish between restrictive disease (in which the lung volumes are decreased: e.g., cystic fibrosis) and obstructive diseases (in which the lung volume is normal but the air flow rate is impeded; e.g., emphysema).)
Alteration of the relationship between physician and patient
The introduction of the techniques of percussion and auscultation into medical practice altered the relationship between physician and patient in a very significant way, specifically because these techniques relied almost entirely upon the physician listening to the sounds of the patients body.
Not only did this development greatly reduce the patient's capacity to observe and contribute to the process of diagnosis, it also meant that the patient was often instructed to stop talking, and remain silent.
As these sorts of evolutionary changes continued to take place in medical practice, it was increasingly necessary to uniquely identify data that was accessible only to the physician, and to be able to differentiate those observations from others that were also available to the patient, and it just seemed natural to use "signs" for the class of physician-specific data, and "symptoms" for the class of observations available to the patient.
King proposes a more advanced notion; namely, that a sign is something that has meaning, regardless of whether it is observed by the physician or reported by the patient:
The belief that a symptom is a subjective report of the patient, while a sign is something that the physician elicits, is a 20th-century product that contravenes the usage of two thousand years of medicine. In practice, now as always, the physician makes his judgments from the information that he gathers. The modern usage of signs and symptoms emphasizes merely the source of the information, which is not really too important. Far more important is the use that the information serves. If the data, however derived, lead to some inferences and go beyond themselves, those data are signs. If, however, the data remain as mere observations without interpretation, they are symptoms, regardless of their source. Symptoms become signs when they lead to an interpretation. The distinction between information and inference underlies all medical thinking and should be preserved.:89
In some senses, the process of diagnosis is always a matter of assessing the likelihood that a given condition is present in the patient. In a patient who presents with haemoptysis (coughing up blood), the haemoptysis is very much more likely to be caused by respiratory disease than by the patient having broken their toe. Each question in the history taking allows the medical practitioner to narrow down their view of the cause of the symptom, testing and building up their hypotheses as they go along.
Examination, which is essentially looking for clinical signs, allows the medical practitioner to see if there is evidence in the patient's body to support their hypotheses about the disease that might be present.
A patient who has given a good story to support a diagnosis of tuberculosis might be found, on examination, to show signs that lead the practitioner away from that diagnosis and more towards sarcoidosis, for example. Examination for signs tests the practitioner's hypotheses, and each time a sign is found that supports a given diagnosis, that diagnosis becomes more likely.
Special tests (blood tests, radiology, scans, a biopsy, etc.) also allow a hypothesis to be tested. These special tests are also said to show signs in a clinical sense. Again, a test can be considered pathognonomic for a given disease, but in that case the test is generally said to be "diagnostic" of that disease rather than pathognonomic. An example would be a history of a fall from a height, followed by a lot of pain in the leg. The signs (a swollen, tender, distorted lower leg) are only very strongly suggestive of a fracture; it might not actually be broken, and even if it is, the particular kind of fracture and its degree of dislocation need to be known, so the practitioner orders an x-ray. The x-ray film shows a fractured tibia, so the film is said to be diagnostic of the fracture.
^Stubbe, H. (Henry Stubbes), The Plus Ultra reduced to a Non Plus: Or, A Specimen of some Animadversions upon the Plus Ultra of Mr. Glanvill, wherein sundry Errors of some Virtuosi are discovered, the Credit of the Aristotelians in part Re-advanced; and Enquiries made..., (London), 1670, p. 75