Cutaneous malignant melanoma is a type of skin cancer, which leads to more than 8000 deaths per year in the United States (Wortsman et al., 2012). By next year every 1 of 50 people is expected to have melanoma. The incidence of melanoma worldwide shows a highly growing tendency including the United States, where its rate of increase has surpassed that of any other cancer (Rigel et al., 2010).
When somebody is diagnosed with melanoma, the lesion is usually removed by a surgeon. However, if the melanoma is not diagnosed in time and moves deep into the dermis, it is likely to break of and spread. These cells enter the lymphatic system and tend to reach the lymph nodes, from where they can reach every part of the body causing serious damage (Kaufman et al., 2005). In some Eastern European countries, including Hungary, patients with skin lesions may face serious limitation when they try to access special health care services because of the lack of dermatologist and trained clinicians with experience in the differential diagnostic methods of skin cancers. However, the early detection and the elimination of skin lesions by a surgeon is critically important in order to avoid the serious consequences associated with the melanoma. When somebody has a chance to have cancer, the world that he/she does not want to hear is: maybe or 'you have to wait for weeks.' People want a quick and accurate diagnosis.
However, the diagnosis (by dermatologists) and subsequent therapy (by surgeons) of a skin lesion is made difficult by the uncertainty of its geometric extent and sometimes even type. Visual inspection is often insufficient as it does not provide enough information. Ultrasound in the last few decades has grown to become one of the most widely used non-invasive diagnostic method in medicine. Ultrasonography for early detection of malignant melanoma is a valuable diagnostic tool. This modality complements the traditional (for example dermoscopic) skin lesion measurements by adding depth and further morphological information. Application of high-frequency ultrasonography for the measurement of skin thickness dates back to 1970's. The first researchers who have shown an A-mode ultrasound device to be a viable tool for measuring skin thickness were Alexander and Miller in 1979 (Jasaitiene et al., 2011). In order meet the widespread diagnostic needs several investigations have been carried out in the last 30 years to determine the most appropriate ultrasound parameters for melanoma detection (Jasaitiene et al., 2011).
On the other hand, the widespread acceptance of superficial ultrasound imaging is hindered by the lack of experience with ultrasound on the part of non-radiologist clinicians, the difficulty of interpretation, as well as the lack of portability and cost of some diagnostic systems.
Out aim is to help both people suffering from skin lesions both untrained clinicians to get a precise diagnosis as quickly as possible. We are hoping, that our ultrasound-based device can lower the chance of the development of consequences associated with melanoma.
The structure of human skin and different skin lesions
In order to understand the morphology of skin lesions and the diagnostic methods, it is important to have some basic information about the structure of our biggest organ, the skin.
The human skin serves a physical barrier between the body and the environment. It protects form the outer space while provides a continuous interaction with it. The skin plays an important role in thermo-regulation, sensation, and protection from ultraviolet radiation by sun, has an important role in D Vitamin regulation and provides a physical appearance. The epidermis, the dermis and the subcutis are the three basic structural layers that form the skin (Bouwstra et al., 2002; Leroy et al., 2013). The epidermis consists of several layers, including stratum corneum, stratum lucidum, stratum granulosum, stratum spinosum, and stratum basale (Kaufman et al., 2005). The outermost layer of the skin is formed by the stratum corneum. This nonviable layer consists of 'brick and mortal' cells and is continuously renewed by the viable - "live"- epidermis (Leroy et al., 2013). Cells from the stratum basale tend to move to the surface forming the stratum corneum. Stratum basale also contains melanocytes, which produce the pigment melanin. The number of melanocytes determins the color of the skin. Melanin helps in protection against harmful ultraviolet radiation coming from the sun (Kaufman et al., 2005).
The dermis can be divided into two additional layers: the papillary dermis, which keeps dermis and epidermis together, and the reticular dermis. The reticular dermis has an important role in thermoregulation and providing nutrients to the upper layers, considering that it contains blood vessels and sweat glands. This layer of the dermis also contains nerve endings, sebaceous and hair follicles (Kaufman et al., 2005).
The hypodermis is an important thermoregulatory, mechanical supporter and nutrition bank for the skin and the whole body.
Our skin, the biggest organ of our body, is vulnerable. There may be several lesions on the surface of the skin. One of them is melanoma. According to The Melanoma Book (Kaufman et al., 2005) ''melanoma is a type of skin cancer that arises from the melanocytes, one particular type of skin cell. Melanoma occurs when the melanocytes become severely abnormal, start dividing uncontrollably, and eventually spread to other parts of the body, sometimes resulting in death.' There are many different types of skin cancer and there are many different classification systems to distinguish them: they can be classified as benign/malignant cancers or pigmented/non-pigmented skin cancers (Argenziano et al., 2003; Braun et al., 2005).
The morphology of the tumor highly depends on in which layer of the skin it is located, take for example its color; the black color of melanocytic neoplasm indicates that it is located in the upper layers of epidermis, particularly in the stratum corneum. Tumors located in the deeper layers of the epidermis tend to appear light to dark brown, and lesions that are in the dermis have a bluish hue, combined with gray in the papillary dermis, and steel-blue in the reticular dermis (Braun et al., 2005). Tumors show an increased nutrition and oxygen intake, which is supported by dilatation of blood vessels or neovascularization which lends a red color for the lesion (Braun et al., 2005). Some tumors may show a white color resulting from scarring or regression (Braun et al., 2005).
Not all melanoma have a color. Those melanomas that do not contain any pigment are called amelanotic melanomas (Kaufman et al., 2005).
Benignant lesions contain epidermal cysts, trichilemmal cysts (TC), pilonidal cysts (PC), pilomatrixoma, dermatofibroma, vascular anomalies, hemagiomas of infancy and vascular malformations (Wortsman et al., 2013) .The most common benign skin lesions are different kind of moles, including common moles, blue moles, atypical or dysplastic moles, congenital nevus and Spitz moles.
Malignant skin lesions can be separated into two groups: nonmelanoma skin cancer (NMSC) and malignant melanoma (MM). NMSCs are the most common types of human skin cancer. Nonmelanoma skin cancer involves basal cell carcinoma and squamous cell carcinoma (Wortsman et al., 2013). Basal cell carcinoma is the most common skin cancer, mainly present on the skin highly exposed to sun, such as face and scalp (Kaufman et al., 2005). The incidence of basal cell carcinoma increases with aging and high exposure to sun. Squamous cell carcinoma is the result of the abnormal division of keratin-producing cells (Kaufman et al., 2005). Similarly to basal cell carcinoma, this type of cancer is more likely to occur on the parts of the body that are highly exposed to sun.
Malignant melanoma constitutes 4-11% of skin cancers (Wortsman et al., 2013). Melanoma can occur on any part of the body where melanocytes are present (Kaufman et al., 2005). Melanoma that appears on the skin is known as cutaneous melanoma and they give 92% of all of the melanomas (Wortsman et al., 2013; Kaufman et al., 2005). (Other types of melanoma are mucosal melanoma 2%, observed around the mouth and nasal cavities, anal melanomas, and ocular melanomas 5% (Kaufman et al., 2005)).
Skin cancer diagnosis
Traditional diagnosis of melanoma highly relies on clinical and histological features of the disease (Wortsman et al., 2013). These features in most of the cases are determined by visual inspection.
Total body photography allows the photographic documentation of high-risk patients or large number of moles, in this way creating a personal dermatologic history. During the examination a trained clinician takes photos from the whole body of the patients, and determines the different parameters of the nevus (S. Q. Wang et al., 2010; D. S. Rigel et al., 2010).
The dermoscope is a hand-held magnifier equipped with a light source used to investigate abnormalities on the skin. (Wang et al., 2010; Rigel et al., 2010) Some devices use oil or alcohol in order to provide a media, which decreases the diffraction, reflection and refraction of light (Wang et al., 2010). The use of dermoscopes in clinical practice significantly improved the specificity and sensitivity of melanoma diagnosis (Rigel et al., 2010). Sensitivity defines the rate of malign melanomas identified as malign melanoma, while specificity means the rate of non-melanomas classified as non-melanoma. In addition, dermoscopes provide an improved diagnostic confidence, while decreases the number of unnecessary biopsies and elimination of lesion (Wang et al., 2010). In the last few years a lot of investigations took place in order to improve both the specificity and sensitivity of differential diagnosis of melanoma.
New dermoscopes compared to their predecessors use special cross-polarizing filters in order to replace the oil and alcohol needed to reduce artifacts (Rigel et al., 2010).
Several diagnostic algorithms have been developed in order to help clinicians in their decisions based on visual inspections. A group of these algorithms are containing score based algorithms, such as the ABCD acronym (Asymmetry, Border irregularity, Color Variegation, Diameter >6mm) developed in 1985 by Rigel et al. The ABCD acronym aims to help clinicians to notice the most common features of melanoma in early stages (Wang et al., 2010; D. S. Rigel et al., 2010).
The CASH system (Color, Architecture, Symmetry, and Homogeneity) through adding information about the architecture of the tumor provides an additional feature in melanoma diagnosis (Henning et al., 2007). Malignant melanomas differ widely from benign melanocytes in terms of color, shape, homogeneity and symmetry of pattern, which may serve as a basis of the CASH method (Braun et al., 2005). Further algorithms are the Seven-point checklist or gestalt-based methods, which compared to the ABCD rules, are based on the pattern analysis of the tumor (Argenziano el al., 1998). According to Argenziano et al. the seven-point checklist gives a score to each of the skin irregularities base on major (atypical pigment network, blue-white veil, atypical vascular pattern) and minor criteria (irregular streaks, irregular pigmentation, dots and globules, regression structures) (Braun et al., 2005; Argenziano el al., 1998).
Mezies et al. developed another alternative for melanoma diagnosis in 1996 (Argenziano et al., 2007; Braun et al., 2005). The Menzies method relies on identifying nine key features in support of, or two features in opposition of a melanoma diagnosis (Menzies et al., 1996).
Several computer assisted diagnostic modalities have been improved in order to increase the specificity and sensitivity of traditional diagnostic methods. Computer based algorithms aim to assist clinicians to distinguish benign and malignant melanoma with minimal involvement (Wang et al., 2010). Computational processing of imaging data involves a pipeline of different operations (Wang et al., 2010). The first step is image acquisition, utilizing either classical digital photographs or specialized dermoscopic imaging methods. Subsequently, image segmentation separates the part of the image with the lesion from the background, and then feature extraction describes the lesion with a number of quantitative parameters, which helps in the differential diagnosis of the tumor. These factors are weighted and summed for comparison with a prescribed threshold. This process serves to quantitatively describe a lesion, and to identify the most relevant features that help to differentiate between benign and malign skin tumors (Wang et al., 2010). Such computer-assisted systems are MelaFind (MELA Sciences, Inc, Irvington, NY, sensitivity of 100%, specificity of 85%) or SIA scope (Biocompaibles, Surrey, UK, 82.7% sensitivity, 80% specificity) (Wang et al., 2010). There any many problems associated with computer based diagnostic methods for the differential diagnosis of melanoma. A common problem associated with computer-assisted method is that they are based on clinically difficult skin lesions (which is an input of neural networks they use), which means that they may fail to recognize the difference between malign and benign tumors when non-expert clinicians use them. Secondly, these methods do not examine the lesion in the context of the neighboring lesions, which leads to a significant loss of important information (Wang et al., 2010).
Confocal scanning microscopy is a non-invasive in and ex vivo technology to improve the sensitivity and specificity of melanoma detection. This technology enables the examination of the skin in a resolution, which is close to histology. However, it is important to note the high cost and lack of portability, and the difficulty of interpretation of images that are associated with this technique which makes its dermal application difficult and not widely accepted (Branzan et al., 2007). RT-PCR based analyzing methods provide the possibility to study tumors with high sensitivity and specificity. Each tumor has its own marker. Reverse transcription of RNA and real time polymerase chain reaction enables the detection of specific hallmarks associated with melanoma and may have an important role in the detection of melanoma without taking biopsies (Keilholz et al., 1998). If the doctor is not concerned enough about the skin lesion just using any of the visual inspection methods listed above, he/she recommends a biopsy or the elimination of the some or the whole critical region. In this cases microscopic evaluations are performed to detect the cancer cells. If the skin cancer is benign, it can be treated by shave biopsy, freezing or laser therapy. In the case of small lesions punch biopsy can be performed with a plastic handler or circular metal core (Kaufman, 2005). If the extent of skin lesion exceeds 4 millimeters, the skin lesion is eliminated by excisional biopsy (Kaufman, 2005). If the skin lesion affects a very large are of the skin, doctors recommend to take just a small biopsy in order to investigate if the lesion should be eliminated or not. This can be performed either by punch biopsy or incisional biopsy (Kaufman, 2005). Taking biopsy may be risky compared to the other methods. As a side effect bleeding and infection can occur.
Another invasive technique to diagnose melanoma is called fine needle aspiration, which is the insertion and quick removal of needles into the melanoma. The cells from the needles are placed onto a glass slide and are examined by a specialist, called a cytologist. The fine needle technique is not a 100% accurate; in some cases the samples taken from the moles may not contain tumor cells, however the cancer is malignant. Secondly, this is just a binary technique: doctors can only decide whether the tumor is malignant or benignant, but cannot obtain any other information regarding the depth or the morphology of the tumor (Kaufman, 2005).
Ultrasound: basics, applications and its role in dermatology
Basics of Ultrasound
Ultrasound is an acoustic wave with a frequency above 20 kHz, which is above the audible range of human hearing. The ultrasound signal is generated by a transducer, which consists of multiple piezoelectric elements, which vibrate by an applied electric current. Transducers are also able to receive acoustic (mechanical) signals from the media and transform them into electric signals as an invers piezoelectric effect.
Ultrasound waves propagate through the target tissue causing vibration disturbance in it. The tissue scatters the ultrasound waves. The properties of scattering highly depend on the physical properties of the tissue, such as its stiffness, density, or water content. The most widely used B-mode (Brightness-mode) imaging method means the transmission of small ultrasound pulses using a transducer (Narouze et al., 2011; Fakoya et al., 2013). The emitted ultrasound waves travel through the tissue. Different components of the tissue have different acoustic impedances, which means that they scatter the ultrasound in different way. The greater the difference between the acoustic impedances of the two layers the higher the amount of reflected ultrasound and the less the amount of transmitted signal. The acoustic impedance (Z) is the product of the mass density of the medium (??) and the speed of propagation (c) according to the following equation
The reflected signals are known as echoes. These echo signals are registered by the same piezoelectric transducer are used to generate an image of the target area (Kost et al., 2006).
Each ultrasound wave can be described by its amplitude, frequency and wavelength. The relation between frequency, sound speed and wavelength can be described by the following equation,
where c denotes the speed of sound in tissue, f is the frequency and ?? is the wavelength.
The ultrasound waves not reflected may cause damage in the tissue (Narouze et al., 2010). The mechanical and thermal indexes aim to describe the damaging potential of ultrasound by providing a guide to the clinicians to the safe application of ultrasound in diagnostics. The thermal index reflects the temperature increase in tissue, while the mechanical index aims to describe the chance of cavitation inside the tissue. Cavitation means the formation, motion and sometimes violent collapse of acoustic bubbles in a media exposed to ultrasound. Cavitation, besides having benefits, may cause serious damages in the tissue. Although the relationship between cavitation and ultrasound intensity is non-trivial, it is well known that there exists a threshold level on intensity below the cavitation does not occur. The ultrasound intensity at a certain depth can be described by the following equation using the absorption coefficient (a), which reflects the amount of absorbed ultrasound (Kost et al., 2006).
I(x) is the ultrasound intensity at depth x, I_0 is the intensity at the surface and a is the absorption coefficient.
Ultrasound waves based on frequency range and application can be divided into three
main groups (Kost et al., 2006):
(1.) High-frequency or diagnostic ultrasound in clinical imaging (3 ' 10 MHz)
(2.) Medium-frequency or therapeutic ultrasound in physical therapy (0.7 ' 3 MHz)
(3.) Low-frequency or power ultrasound in therapy (18 ' 100 MHz)
Therapeutic applications of ultrasound
Ultrasound, a well-known diagnostic tool, seems to have received a growing interest in therapeutics as well. Therapeutic applications of ultrasound are widespread. High intensity focused ultrasound can be used in ophthalmology for treatment of glaucoma, which is a common ocular disease arising from the progressive death of retinal ganglion cells (Aptel et al., 2012). There are several risk factors, which increase the chance of the development of the disease. One of them is the increased intraocular pressure. HIFU based on its physical parameters has the potential to disrupt the ciliary epithelium and to enhance the outflow of aqueous humour through the thinned sclera (Aptel et al., 2012). By the same mechanism, HIFU can be used to disrupt Blood Brain Barrier (BBB), which makes the delivery of different drugs into the brain difficult (Shahrzad et al., 2010). Ultrasound has been shown as a possible method for transdermal drug delivery through the mechanism called cavitation. Ultrasound may facilitate gene therapy and treat vascular thrombosis through the cavitation-based dissolution of blood clots. HIFU has a significant role in cancer therapy as well, by localized activation of encapsulated drugs (Mitragotri et al., 2005).
Diagnostic applications of ultrasound
Sonography is a safe, cheap and non-invasive diagnostic tool, both in human and veterinary medicine (Szabo, 2004). Ultrasound has several diagnostic applications. It is widely used in cardiology for imaging the heart and blood flow based on Doppler effect. Sonography is mostly used for prenatal care. Other fields of ultrasound application include the examinations of different organs of the human body, including kidneys, liver, gallbladder, bladder, uterus, ovaries, scrotum, female breast, and pancreas (Szabo, 2004).
The role of ultrasound in dermatology
Ultrasound is the most widely used imaging tool in medicine. In the last few years high-frequency ultrasound seems to have achieved a significant role in preoperative investigation of skin melanoma, too (Rai??utis et al., 2010). High-frequency (>100 MHz) ultrasound has a high axial resolution (~16 ??m), which enables us to visualize the morphological structure of the skin, however the penetration depth is limited (~1.5 mm) (Jasaitiene et al., 2011.). Based on this simple physical consideration, determination of the parameters of the sonication is of high importance. 7.5 ' 20 MHz ultrasonic transducers are the most common tools used to determine skin thickness and density measurements, which can serve as a possible input parameter for a subsequent therapy (Jasaitiene et al., 2011).
As it was described in the previous sections, human skin consists of three major functional units including epidermis, dermis and hypodermis. Based on its special cell types, each of the layers defined above have its own echogenicity profile (Wortsman et al., 2013), which gives a high relevance to the ultrasound based diagnostic methods. The echogenicity of epidermis is highly determined by the korneocytes, the main cell type in the epidermis. The dermis contains a lot of collagen, which gives a basic structure to the skin and affects the sonographic profile of this layer. Last, but not least, the echogenicity of hypodermis comes from the fatty cells and the blood vessels found in this layer. The skin has several annexes. These annexes include for example nails and hair follicles. Echogenicity of nail bed may vary site by site. Hair follicles, just like the nails, are very hypoechoic, and have no significant effect on the ultrasound images (Wortsman et al., 2013).
Tumors also have their own echogenic profiles, as Wortsman et al. described it: 'Histologically, BCC shows islands of atypical basaloid cells and fibrous stroma' On ultrasound images, BCC appears as a well-defined, oval-hasped, hypoechoic lesions that commonly present hyperechoic spots with increased vascularity around the tumor' MMs are composed of atypical melanocytes that present irregular nuclei, nuclear pleomorphism, and marked mitotic activity. MMS tend to show as well-defined, fusiform hypoechoic lesions with prominent vascularity.' (Wortsman et al., 2013)
Color, Power, and Continuous wave Doppler
Doppler ultrasound is able to detect tissue in motion, such as blood flow. Because of this, it is able to monitor the vascularization state of the skin cancer. The uptake of oxygen and nutrition of cancer cells is much higher than in the normal cells. Because of this, cancer cells develop new blood vessels, to support their needs. The development of new blood vessels in tumors is called angiogenesis, which is one of the hallmarks of cancer (Hanahan et al., 2000; Hanahan et al., 2011). These characteristics of the tumors make them much more detectable by different kind of Doppler ultrasounds. The principle of Doppler ultrasound relies on the phenomenon that the transducer and the reflector (in this case the blood cells) of the ultrasound wave are moving relative to each other. Doppler devices are able to detect changes in the initial emitted frequency (Kleinerman et al., 2011).
Color Doppler Sonography has been shown to improve the accuracy of the diagnosis of melanoma and metastasizes compared to the conventional sonographic procedures (Moehre et al., 1999). Color Doppler machines with variable frequency probes make the visualization of echo structure of the superficial skin layers possible (Wortsman et al., 2012; Machet et al., 2011). Cammarota et al. highlighted, that the combination of color and power Doppler with gray scale imaging would enable the investigation of pathologic vascularization state of skin lesion by identifying abnormal flow signals (Cammarota et al., 1998). It has been shown, that power Doppler is able to differentiate melanomas of higher thickness, which is very challenging to single frequency ultrasound because of its limited penetration depth (Wortsman et al., 2012). Although power and color Doppler has the potential of real-time monitoring of blood flow and texture of the lesion, it needs sophisticated technology, which makes its application very expensive. Color Doppler, as a variable frequency machine, requires expensive multichanneled ultrasound hardware capable of operating in the frequency range of 15 ' 22 MHz (Wortsman et al., 2013).
Pulsed wave Doppler is a viable tool to determine the depth of the moving object through the delay between the emitted signal and received signal (Kleinerman et al., 2011).
Continuous Wave Doppler gives real time information of the presence of moving objects, but does not give any information regarding their depth (Kleinerman et al., 2011).
Fixed frequency ultrasound
Application of fixed high frequency probes (20-100 MHz) also resulted in information that showed a significant correlation between sonometry and histometry (Wortsman et al., 2013). However, the biggest issue is the limited penetration depth (5-6 mm at 20MHz, 3 mm at 75 MHz, and 1 mm 100 MHz). High frequency, 20-25 MHz ultrasound is able to capture the epidermis and the dermis, while higher frequency devices (50-100 MHz) only enable the inspection of epidermis (Kleinerman et al., 2012).
Several studies have shown that ultrasound is a viable tool to determine how deeply the tumor cells have invaded (Hoffman et al., 1992; Gambichler et al., 2007). In ultrasound images tumors appear as homogeneously hypoechoic regions surrounding by hypoechoic normal skin regions (Wang et al.,). All of these facilities make possible the accurate diagnosis and the deremination of the depth of melanoma.
Contrast enhanced ultrasound
Contrast enhanced ultrasound is a powerful tool to investigate primary skin lesions and its metastases. Contrast enhanced ultrasound in some studies was able to identify benign and metastatic lymph nodes with a sensitivity of 100% and specificity of 99.5% (Tombesi et al., 2011). Prior to this functional analysis, an ultrasound contrast medium is added to the patient. These contrast agents are usually micro bubbles that increase the signal to noise ratio by increasing the contrast. Small air-filled bubbles inside the vessels backscatter or reflect the ultrasound waves, resulting in a high echogenicity. These simple physical principles make the visualization of blood flow and micro vessel detection of melanoma possible (Wortsman et al., 2013).
Elastography is a technique used to determine the mechanical properties of tissue. The basic mechanism of sonoelastography is very similar to palpation. During the measurements, the tissue is exposed to low-frequency acoustic force, which causes a distortion in the tissue, called strain. The degree of local distortion can be measured, giving a quantitative measure of a particular tissue area. Elastography has several application areas, including the real-time monitoring of HIFU treatment, which causes a local change in the stiffness of tumor due to thermal ablations (dr. Mikl??s Gy??ngy, Diagnostic Ultrasound Imaging Course, Elastography, P??zm??ny P??ter Catholic University). In the last years researchers have shown that elastography has the potential to describe different biological processes, such as aging associated with human skin (Fujimora et al., 2008). It has been also shown that benign tumors have lower elasticity compared to the malign ones, which information can serve as an important parameter in the differential diagnosis of tumors (Kleinerman et al., 2011).
Comparison of different diagnostic methods
Based on the following table, we can conclude that ultrasound has almost 100% specificity and 100% sensitivity. Moreover, unlike other methods, information on the spatial extent is provided. This is crucial, since an inadequately excised melanoma has possibility of recurrence. Even though ultrasound has many advantages over other methods, the main disadvantage associated with ultrasound imagines is that the diagnosis of the lesion requires a trained clinician.
COMPARISION OF EMERGING AND OLD TECHNOLOGIES
TECHNOLOGY SENSI-TIVITY SPE-CIFI-CITY ADVANTAGES DISADVANTAGES
MoleMax N/A N/A Two camera system; no oil immersion required; transparent overlay for follow up; total body photography No computer diagnostic analysis
MelaFind 95 ' 100% 70 ' 85% Multispectral sequence of images created in <3 seconds; Handheld scanner -
Spectophotometric intracutaneous analysis 83 ' 96% 80 ' 87% Diagnosis of lesions as small as 2 mm in diameter; observes skin structure, vascular composition and reticular pigment networks; handheld scanner -
91% 68% Empirical database for comparison; session, and image-level accuracy calibration; recorded on graphic map of body Requires oil immersion
Confocal scanning laser microscopy (CSLM) 98% 98% Histopathological evaluation at bedside with similar criteria; longer wavelengths can measure into papillary dermis; fiber-optic imaging allows for flexible handheld devices Poor resolution of chromatin patterns, nuclear contours and nucleoli; can only assess to depth of 300 ??m; melanomas without in situ component will likely escape detection
Optical coherence tomography (OCT) N/A N/A High resolution cross-sectional images resembling histopathological section of skin; higher resolution than ultrasound and greater detection depth than CSLM Photons are scattered more than once, which can lead to image artifacts; ointment or glycerol may be needed to reduce scattering and increase detection depth; observation of architectural changes and not single cells
Ultrasound technology 99% 99% Cost effective; information about inflammatory processes of skin in relation to nerves and vessels Tumor thickness may be overestimated because of underlying inflammatory infiltrate; melanoma metastasis cannot be separated from that of another tumor; images can be difficult to interpret
Tape stripping mRNA 69% 75% Rapid and easy to perform; painless; practical for any skin surface; can retest same lesion Results based upon small data set delay in getting test results need larger gene expression profiles for comparison
Electrical bioimpedance 92 ' 100% 67 ' 80% Complete examination lasts 7 minutes Electrical impedance properties of human skin vary significantly with body location, age, gender, and season
NOTE: This table was taken from (Rigel et al., 2010)
As it was highlighted, there are many possibilities to use ultrasound for melanoma detection. However, a common problem associated with them is that they are not portable enough for comfortable use for the immediate diagnosis of lesions; 'they are not designed with teleradiology in mind' (dr. Mikl??s Gy??ngy). Our solution for this problem is a highly portable and easy to use device that is able to describe the properties of a lesion for non-expert users of ultrasound with specificity and sensitivity close to 100%. This can be achieved through a device that integrates the image acquisition, processing and analysis in order to deliver a set of quantitative parameters that helps diagnosis.
Our product consists of three distinct but tightly coupled components. The ultrasound module that incorporates the transducer and the driver, the tablet device, that performs image analysis, real time visualization, and serves as a gateway to the third component; the cloud system. Firstly, I will describe the ultrasound module.
Our low-cost portable ultrasound system is driven by a microprocessor, which controls a field-programmable gate array that carries out digital signal processing of the input and output signals to/from the transducer.
Figure 1.1: Overall system block diagram (Based on Gi-duck et al., 2012)
The system is equipped with a wireless communications module that transmits real-time imagery to a tablet, and to a cloud-driven diagnostic system for teleradiological purposes.
The transducer is a single frequency 20 MHz 128 elements transducer. It is connected to the field-programmable gate array (FPGA), which in turn contains ultrasound signal and image processing modules, transmit and a 32-channel receive beamformer with pseudo-dynamic focusing, providing B-mode imaging.
The system does not require external power supply during the examination; its lithium-ion battery lasts up to approximately 2 hours and can be recharged.
The system has various Input/Output connections, both for diagnostic purposes and for transmitting data. Data can be directly exported onto a micro SD card, transmitted wirelessly through Wi-Fi 802.11a/b/g/n or Bluetooth, or via a USB wired connection.
The Tablet and the App
The Android or iOS App provides real-time video imagery about the target area. As a back process, it performs image analysis in order to provide quantitative and qualitative data and help untrained clinicians to perform differential diagnosis. The video feed and the interactive touchscreen can be used to control dynamic focusing, and limited zoom capabilities will be available. Furthermore, various image overlays over the original video feed highlight different regions of interest; thereby assisting trained or even untrained clinicians. The system provides the option to make snapshots, label and catalogue them, and then share them via the internet or to print them. The App helps streamline the diagnostic process; clinicians can take a total body photograph using the tablet, then they can match the further camera and 3 dimensional ultrasound images to certain parts of the body.
The algorithm implemented in either the tablet or cloud will perform advanced image processing and parameter quantification tasks: first of all, image segmenting separates the background from the B-mode image of the lesion. This step involves the detection of hyper- and hypo-echogenic regions, which are the hallmarks of tissue with different density. As it was described, malignant and benign skin lesions have different echogenic profiles, which can serve as an input parameter for the further analysis. Based on this segmentation information the spatial characteristics such as width and depth can be described. All of this information may lead to a final step: the automatic quantitative differential diagnosis of melanoma.
Cloud based teleradiology enables the distant inspection and annotation of transferred images by professionals. Doctors can highlight the relevant regions just using the touchscreen and images can be easily shared with colleagues.
This chapter aimed to give an overview of the main properties of skin lesions and the potential of different technologies used for melanoma detection. I have also introduced our proposed device that aims to address many of the shortcomings of existing solutions: it is a portable, easy to use device, which gives valuable information regarding the morphology of the tumor.
Source: Essay UK - http://www.essay.uk.com/free-essays/science/cutaneous-malignant-melanoma.php