Investigation of dental samples using a 35 MHz focussed ultrasound piezocomposite transducer D.A. Hughes a,e, * , J.M. Girkin a , S. Poland a , C. Longbottom b , T.W. Button c , J. Elgoyhen c , H. Hughes c , C. Meggs c , S. Cochran d a Institute of Photonics, SUPA, University of Strathclyde, Wolfson Building, 106 Rottenrow, Glasgow, Scotland G4 0NW, UK b Centre for Clinical Innovations, University of Dundee, Dundee, Scotland, UK c Applied Functional Materials Ltd., Birmingham Research Park, Birmingham B15 2SQ, UK d Institute for Medical Science and Technology (IMSAT), University of Dundee, Wilson House, 1 Wurzburg Loan, Dundee Medipark, Dundee DD2 1FD, Scotland, UK e Bioengineering Department, University of Strathclyde, Wolfson Building, 106 Rottenrow, Glasgow, Scotland G4 0NW, UK article info Article history: Received 6 April 2008 Received in revised form 13 August 2008 Accepted 16 August 2008 Available online 9 September 2008 Keywords: Human teeth Biomedical imaging High frequency abstract Dental erosion and decay are increasingly prevalent but as yet there is no quantitative monitoring tool. Such a tool would allow earlier diagnosis and treatment and ultimately the prevention of more serious disease and pain. Despite ultrasound having been demonstrated as a method of probing the internal structures of teeth more than 40 years ago, development of a clinical tool has been slow. The aim of the study reported here was to investigate the use of a novel high frequency ultrasound transducer and validate it using a known dental technique. A tooth extracted for clinical reasons was sectioned to provide a sample that contained an enamel and dentine layer such that the enamel–dentine junction (EDJ) was of a varying depth. The sample was then submerged in water and a B-scan recorded using a custom-designed piezocomposite ultrasound trans- ducer with a centre frequency of 35 MHz and a 6 dB bandwidth of 24 MHz. The transducer has an axial resolution of 180 lm and a spatial resolution of 110 lm, a significant advance on previous work using lower frequencies. The depth of the EDJ was measured from the resulting data set and compared to measurements from the sequential grinding and imaging (SGI) method. The B-scan showed that the EDJ was of varying depth. Subsequently, the EDJ measurements were found to have a correlation of 0.89 (p < 0.01) against the SGI measurements. The results indicate that high fre- quency ultrasound is capable of measuring enamel thickness to an accuracy of within 10% of the total enamel thickness, whereas currently there is no clinical tool available to measure enamel thickness. Ó 2008 Elsevier B.V. All rights reserved. 1. Introduction Although ultrasound has been used for diagnostic applications in medicine since the mid 1940s [1], it has not developed into a mature diagnostic technique in dentistry. In the late 1960s it was shown that dental structures could be observed using ultrasound [2]. However, further work has been sporadic despite ultrasound images having been utilised in both the longitudinal and trans- verse direction for probing the inner structures of teeth [3,4]. In this paper a high frequency ultrasonic investigation of dental samples is reported utilising a novel transducer to produce a B- scan. Verification of the B-scan is performed using the sequential grinding and imaging (SGI) method. The enamel of the tooth, while being the hardest material in the human body, is only able to survive in environments down to around pH 4.6. At this point the surrounding environment becomes too acidic and the enamel starts to dissolve [5]. Due to changes in lifestyles and dietary habits, there is an increasing prevalence of acid erosion worldwide [6]. Accurate measurement of tooth wear in vivo over short periods of time is problematic due to lack of fixed points on an eroding tooth surface and lack of optical instrumenta- tion to measure minute changes in vivo [7]. Whilst it is important to: (1) identify patients with erosion early and (2) monitor whether an eroded surface is stabilised or still being eroded, unfortunately the clinical visual signs of early ero- sion are weak. This is because there are no stable definitive clinical reference points (on the tooth surface) between the stages of the earliest initial loss and the complete loss of the full enamel thick- ness [8,9]. Hence, the detection and monitoring of enamel erosion are problematic: currently there is no diagnostic device available for the early detection and quantification of dental erosion [5]. Thus, the motivation of the work reported here is the rising prev- alence of dental erosion and the lack of a diagnostic device which can measure and monitor the disease in a quantitative manner. 0041-624X/$ - see front matter Ó 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.ultras.2008.08.007 * Corresponding author. Address: Institute of Photonics, SUPA, University of Strathclyde, Wolfson Building, 106 Rottenrow, Glasgow, Scotland G4 0NW, UK. Tel.: +44 141 548 4120. E-mail address: david.hughes@strath.ac.uk (D.A. Hughes). Ultrasonics 49 (2009) 212–218 Contents lists available at ScienceDirect Ultrasonics journal homepage: www.elsevier.com/locate/ultras