Caries Diagnosis: An Introduction


Diagnosis vs DetectionClassical Diagnostic Decision Process


Diagnosis is 'the art of identifying a disease from its signs and symptoms' (Merriam Webster, 2003). It is the amalgamation of several factors; a complex process involving pattern recognition, probability, deductive thinking, and careful consideration of the circumstances of the individual patient. It is a deductive and evaluative thought process. Diagnostic decision-making is thus a delicate balancing act, and the clinician who makes such a decision has to evaluate the various factors based on the signs and symptoms surfaced by the patient, and his own clinical experience and training. Diagnosis is therefore more of an 'art' than an 'exact science' (Nyvad, 2004).

Detection, on the other hand, refers to the mere recognition of the signs and symptoms based on a given set of principles. It is empirical rather than deductive, an 'exact science' rather than an 'art'. To detect a disease, therefore, suggests a classifying or labelling of a patient based on a set of fixed parameters and tests.

It is crucial to note that the systems and methods described in this project would be more accurately classified as detection systems rather than diagnostic systems. These detection systems are thus aimed at augmenting the diagnostic process and facilitate the clinician's decision-making by enabling either early detection of the disease, or presenting the disease in a quantifiable manner. The actual diagnosis rests on the shoulders of the clinician, who will use the information provided by these detection systems as an aid in reaching his diagnosis, but he will also not discount the other contributing factors (such as patient history, diet, stastical probability etc) in order to reach a balanced diagnosis and provide a sound treatment plan. (Pretty, 2006)

(Nyvad, 2004)

Statistical Concepts in Diagnostic Test Systems

In most diagnostic tests, the test results tend to follow the distribution as shown in the graph below.

Graph showing typical distribution of diagnostic test results

(Graph taken from
http://www.medcalc.be/manual/roc.php)

Of the two populations which undergo the diagnostic test, where one population has the disease, and the other does not, the results show that there is rarely a perfect separation between the two groups. This is due to certain inaccuracies inherent in the diagnostic test used. For every criterion value used to discriminate between the two populations, there will be regions which correspond to correct diagnoses and incorrect diagnoses. In some cases, the disease is correctly classified as positive (TP = True Positive), but in some cases, the disease is incorrectly classified as negative (FN = False Negative). On the other hand, the cases without the disease may be correctly classified as negative (TN = True Negative), or incorrectly classfied as positive (FP = False Positive).

The different fractions (TP, FN, TN, FP) may be represented in the following table:


Disease





Test Present n
Absent n
Total
Positive True Positive (TP) a
False Positive (FP) c
a+c
Negative False Negative (FN) b
True Negative (TN) d
b+d
Total
a+b

c+d


(Table taken from
http://www.medcalc.be/manual/roc.php)

Some of the statistical concepts used commonly in diagnostic testing are defined as such:

Sensitivity: The probability that a test result will be positive when the disease is present [a / (a + b)]
Specificity: The probability that a test result will be negative when the disease is absent [d / (c + d)]

These two concepts are often used to determine the usefulness and effectiveness of diagnostic tests, including the caries diagnostic methods which will be discussed in this project.


An Overview of Caries Detection Methods


Methods of caries detection and diagnosis can generally be classified into two kinds, namely the conventional methods that have been around for a long time, (such as visual inspection, visual-tactile diagnosis, and conventional radiography), and the novel methods which have been more recently introduced (such as laser fluorescence, fiber-optic transmission, digital radiography, and electrical methods).

While conventional methods have been employed extensively because of their ease of use, they lack sensitivity and specificity, and are only able to detect the presence of caries at a later stage. With the increasing standard of healthcare and improvements in technology, novel methods have been introduced which aim to detect caries at a much earlier stage than possible with traditional methods. These methods make use of the measurement of a particular physical signal - ranging from X-rays, visible light
, laser light, electrical current, ultrasound - and the device is able to receive and interpret the signals in a meaningful way that can be quantified and used as a diagnostic tool. The following diagnostic methods will be discussed in this project:


Visual-Tactile Caries Diagnosis


Radiography


Fibre Optic Transillumination


Electrical Conductance


Laser Fluorescence


Towards The Future: Other New and Emerging Technologies



Quick Links
Homepage| Contents Page | Introduction | Visual-Tactile Caries Diagnosis | Radiography| Fiber-Optic Transillumination (FOTI)
Electrical Conductance | Laser Fluorescence And QLF | Towards The Future: Other New And Emerging Technologies | A Summary


_________________________________________________________________________________________________________________________________________

References

1) Nyvad, B. (2004). Diagnosis versus detection of caries. Caries Res, 38(3), 192-198.

2. Pretty, I. A. (2006). Caries detection and diagnosis: novel technologies. J Dent, 34(10), 727-739.

3) ROC Curve Analysis: Introduction. Taken from http://www.medcalc.be/manual/roc.php.



More pages