EXPERT INTERPRETATION COMPENSATES FOR REDUCED IMAGE QUALITY OF CAMERA‐DIGITIZED IMAGES REFERRED TO RADIOLOGISTS

We compared the accuracy of five veterinary radiologists when reading 20 radiographic cases on both analog film and in camera‐digitized format. In addition, we compared the ability of five veterinary radiologists vs. 10 private practice veterinarians to interpret the analog images. Interpretation accuracy was compared using receiver operating characteristic curve analysis. Veterinary radiologists' accuracy did not significantly differ between analog vs. camera‐digitized images (P=0.13) although sensitivity was higher for analog images. Radiologists' interpretation of both digital and analog images was significantly better compared with the private veterinarians (P<0.05).     

Comparison of Seven Digital Cameras for Digitizing Radiographs

The purpose of this study was to evaluate seven digital cameras for their ability to adequately capture quality radiographic images and their relative cost effectiveness. Radiographs of line pair (LP) phantoms (ranging from 0.6 to 16.6 LP/mm) and a 20-step contrast phantom were photographed at a near distance of 30.0 cm and a far distance (determined for each camera) at which a full 14 x 17 in film filled the viewfinder. Of the seven cameras tested, the Canon EOS-D30 consistently performed the best in LP/mm resolved (7.7 at close distance, 1.4 at far distance) and in contrast latitude (all 20 steps at both close and far distances). The Canon EOS-D30, determined to be the best camera used in this study for photographing radiographic images, is also the most expensive camera that was used. Two cameras, the Nikon Coolpix 995 and the Sony DSC-F707, were very close to the Canon in performance at considerably less cost. Certain inexpensive cameras may not be acceptable choices for teleradiology.     

Comparative evaluation of six different body regions of the dog using analog and digital radiography

In this study the quality of digital and analog radiography in dogs was compared. For this purpose, three conventional radiographs (varying in exposure) and three digital radiographs (varying in MUSI-contrast [MUSI = MUlti Scale Image Contrast], the main post-processing parameter) of six different body regions of the dog were evaluated (thorax, abdomen, skull, femur, hip joints, elbow). The quality of the radiographs was evaluated by eight veterinary specialists familiar with radiographic images using a questionnaire based on details of each body region significant in obtaining a radiographic diagnosis. In the first part of the study the overall quality of the radiographs was evaluated. Within one region, 89.5% (43/48) chose a digital radiograph as the best image. Divided into analog and digital groups, the digital image with the highest MUSI-contrast was most often considered the best, while the analog image considered the best varied between the one with the medium and the one with the longest exposure time. In the second part of the study, each image was rated for the visibility of specific, diagnostically important details. After summarisation of the scores for each criterion, divided into analog and digital imaging, the digital images were rated considerably superior to conventional images. The results of image comparison revealed that digital radiographs showed better image detail than radiographs taken with the analog technique in all six areas of the body.     

COMPARISON OF COMPUTED RADIOGRAPHY AND CONVENTIONAL FILM-SCREEN RADIOGRAPHY OF THE EQUINE STIFLE

Major advantages of computed radiography are the potential reduction of dose and the possibility of postprocessing. In our study, we compared conventional radiographs to digital radiographs of the equine stifle by subjective evaluation of diagnostic quality when using a decreasing photon flux (mAs). Twelve equine stifle joints from horses of different weight and size were examined. Conventional and digital radiographs were performed identically in a caudocranial projection with the tube angled 15 degrees. A series of four radiographs was performed in each technique with an increasing photon flux starting with 2.5 mAs and going up to 5, 10, and 20 mAs. All radiographs were evaluated subjectively in a blinded fashion by seven readers in terms of contrast, bone structure, and diagnostic value and were graded using a 1-5 scale. Results from conventional and digitized radiographs were compared, and differences between the individual observers were analyzed statistically. Contrast, bone structure, and diagnostic value from digital images were rated significantly better than from conventional images (p < .001). For both techniques, a decrease in ranking was found with a decrease of photon flux. There was only slight interobserver variability. A dose reduction up to a factor of 4 compared to a 100 speed film-screen system seems to be possible without loss of information. Weight and size of the horse are not major influences.     

Comparison of digitized and conventional radiographic images for assessment of hip joint conformation in dogs

OBJECTIVE: To determine agreement between assessments of canine hip joint conformation provided by board-certified radiologists after evaluation of digitized and conventional radiographic images. SAMPLE POPULATION: 200 pelvic radiographs previously evaluated by radiologists using the Orthopedic Foundation for Animals standard grading system for canine hip joint conformation. PROCEDURES: Each of 20 board-certified radiologists evaluated conventional and digitized pelvic radiographs from each of 200 dogs for hip joint conformation. A weighted kappa coefficient and intraclass correlation were used to determine agreement between assessments derived from digitized radiographic images and conventional radiographs and between the original Orthopedic Foundation for Animals conformation ratings and assessments derived from each image format. RESULTS: Overall, agreement between assessments derived from the digitized images and conventional radiographs was good, with all but 1 radiologist attaining a weighted kappa coefficient > 0.61. Intraclass correlation for each radiologist ranged from 0.75 to 0.98 (95% confidence interval, 0.67 to 0.984). On comparison of conformation assessments, 95.7% of those derived from conventional radiographs and digitized images were within 1 grade. On comparison of digitized radiographic conformation assessments and conformation ratings, 94.2% were within 1 grade. On comparison of conventional radiographic conformation assessments and conformation ratings, 92.3% were within 1 grade. CONCLUSIONS AND CLINICAL RELEVANCE: The use of digitized radiographic images does not appear to impact the radiographic assessments of canine hip joint conformation made by consultant radiologists. Compared with conventional radiographs, the use of digitized radiographic images decreases storage space requirements and enables more rapid reporting of assessment results for individual dogs.     

Comparison of radiographic assessments of the tibial plateau slope in dogs

OBJECTIVES: To evaluate the accuracy of 2 radiographic methods used to assess tibial plateau slope (TPS) in dogs and evaluate effects of film digitization and radiographic beam placement on TPS measurements. SAMPLE POPULATION: 16 hind limbs from dog cadavers weighing > 20 kg. PROCEDURES: Radiographs of tibiae were made with the radiographic beam centered over the stifle joint and midshaft of the tibia. Tibiae were collected, the femorotibial contact area was determined, and slope of the medial tibial condyle in relation to the tibial shaft was measured. Radiographs were digitized. Slope of the medial tibial condyle was measured on printed and digitized radiographs read in random order by 6 examiners unaware of anatomic measurements. Three examiners used a conventional measuring technique, and 3 examiners used an alternative measuring technique. RESULTS: Anatomic measurements were significantly higher than radiographic measurements made by use of the conventional interpretation method but did not differ from radiographic measurements made by use of the alternate method. Measurements from printed radiographs were lower than measurements from digitized radiographs for the 4 most experienced examiners. CONCLUSIONS AND CLINICAL RELEVANCE: Measurements made by use of a line tangential to the cranial, linear portion of the medial tibial condyle at the femorotibial contact point were accurate measurements of the anatomic TPS. Measurements made by use of the conventional TPS measurement method underestimated the anatomic TPS. Measurements made on digitized radiographs were typically more accurate than measurements made on printed radiographs.     

Evaluation of Intra‐ and Interobserver Variability and Repeatability of Tibial Plateau Angle Measurements with Digital Radiography Using a Novel Digital Radiographic Program

Objective— To compare the intra‐ and interobserver variability occurring when observers with differing experience levels measure tibial plateau angles (TPAs) with a novel digital radiographic projection program (tibial plateau leveling osteotomy [TPLO] planning program), the Kodak Picture Archiving and Communications System (PACS), and standard sized printed films (SF). Study Design— Cross‐sectional study. Sample Population— Dogs (n=36) with cranial cruciate ligament (CCL) rupture that had a TPLO. Methods— Six observers, divided into 3 equal groups based on experience level, measured TPA on 36 digitally captured radiographic images of tibiae of dogs clinically affected with CCL rupture. Each observer used 3 methods of measuring TPA and repeated the measurements 3 times with each method. The intra‐ and interobserver variability was compared using the coefficient of variation. Results— Averaged over all replications and images, there was no significant difference (P>.05) in the average variability occurring with each method for all but 1 observer. There was no effect of experience level on measurement variability; however, interobserver variability was significantly less with measurements made with the PACS and TPLO planning program compared with measurements made from SF (P<.05). Conclusions— Repeated measurements of TPA made using digital images and computer‐based measurement programs were significantly less variable between observers than those made from images printed on standard radiographic films. Clinical Relevance— Digital radiography and computer‐based measurement programs are effective for determining the TPA, allowing less variability in measurements compared with SF. The ability to manipulate the image may allow better identification of anatomic landmarks.     

Investigation of Ossification Pattern of Structures of Equine Metacarpophalangeal Articulation with the Aid of Digital Imaging Segmentation and Processing Techniques: A Preliminary Report

Radiographs, taken at lateromedial (LM) view, were obtained from metacarpophalangeal articulations of 25 horses of variable ages, ranging from 2 to 20 years, as well as a few of undetermined ages. Images of radiographs were acquired as digitized images with the aid of high-resolution (megapixel) digital cameras directly from lighted platforms, and the acquired 24-bits bitmap formatted images were converted to 8-bit grayscale tagged-image format (TIF) images. The TIF images were processed by a technique involving an on-screen quantitative evaluation of spot pixel intensity values in the digitized images, followed by monadic arithmetic pixel subtraction and addition. A color mapping of the pixels into their respective red and blue sub-spectra of the RGB spectrum also was performed. The entire processing operation was accomplished through the use of a standard digital imaging software. The results demonstrate typical bands of possible ossification pattern within the diaphyses, metaphyses, and the epiphyses of the articulating metacarpal and proximal phalanx in the processed radiographic images compared with their nonprocessed counterparts. The relevance of this novel observation, although preliminary, lies in the prospects of assessing pre-race skeletal health status of horses through a cost-effective, simple, and reliable method.     

ACQUISITION HARDWARE FOR DIGITAL IMAGING

Use of digital radiography is growing rapidly in veterinary medicine. Two basic digital imaging systems are available, computed radiography (CR) and direct digital radiography (DDR). Computed radiographic detectors use a two-step process for image capture and processing. Image capture is by X-ray sensitive phosphors in the image plate. The image plate reader transforms the latent phosphor image to light photons that are converted to an analog electrical signal. An analog to digital converter is used to digitize the electrical signal before computer analysis. Direct digital detectors provide digital data by direct readout after image capture—a reader unnecessary. Types of DDR detectors are flat panel detectors and charge coupled device (CCD) detectors. Flat panel detectors are composed of layers of semiconductors for image capture with transistor and microscopic circuitry embedded in a pixel array. Direct converting flat panel detectors convert incident X-rays directly into electrical charges. Indirect detectors convert X-rays to visible light, then to electrical charges. All flat panel detectors send a digitized electrical signal to a computer using a direct link. Charge coupled device detectors have a small chip similar to those used in digital cameras. A scintillator first converts X-rays to a light signal that is minified by an optical system before reaching the chip. The chip sends a digital signal directly to a computer. Both CR and DDR provide quality diagnostic images. CR is a mature technology while DDR is an emerging technology.     

ACQUISITION HARDWARE FOR DIGITAL IMAGING

Use of digital radiography is growing rapidly in veterinary medicine. Two basic digital imaging systems are available, computed radiography (CR) and direct digital radiography (DDR). Computed radiographic detectors use a two‐step process for image capture and processing. Image capture is by X‐ray sensitive phosphors in the image plate. The image plate reader transforms the latent phosphor image to light photons that are converted to an analog electrical signal. An analog to digital converter is used to digitize the electrical signal before computer analysis. Direct digital detectors provide digital data by direct readout after image capture—a reader unnecessary. Types of DDR detectors are flat panel detectors and charge coupled device (CCD) detectors. Flat panel detectors are composed of layers of semiconductors for image capture with transistor and microscopic circuitry embedded in a pixel array. Direct converting flat panel detectors convert incident X‐rays directly into electrical charges. Indirect detectors convert X‐rays to visible light, then to electrical charges. All flat panel detectors send a digitized electrical signal to a computer using a direct link. Charge coupled device detectors have a small chip similar to those used in digital cameras. A scintillator first converts X‐rays to a light signal that is minified by an optical system before reaching the chip. The chip sends a digital signal directly to a computer. Both CR and DDR provide quality diagnostic images. CR is a mature technology while DDR is an emerging technology.     

ACCURACY OF RADIOGRAPHIC DETECTION OF THE CRANIAL MARGIN OF THE DORSAL LAMINA OF THE CANINE SACRUM

An elongated sacral lamina has been described as one of the contributing factors for dogs with cauda equina syndrome due to degenerative lumbosacral stenosis (DLSS); however, published evidence is lacking on the accuracy of radiographic screening for the presence of this lesion. Objectives of this prospective, cross‐sectional cadaver study were to describe the accuracy and repeatability of detection of the cranial sacral lamina margin on plain lateral radiographs of the lumbosacral junction in dogs. Twenty‐five medium and large breed canine cadavers were radiographed before and after placement of a radiopaque hook in the cranial margin of the sacral lamina. Three independent evaluators placed digital markers at the perceived margin on preinterventional radiographs. The distance from perceived location to the true location on postinterventional radiographs was recorded for each dog and observer. A discordance threshold (distance between perceived and actual margin) of 1.5 mm was subjectively defined as clinically relevant. The three evaluators demonstrated good repeatability, although the accuracy for margin detection was only fair (mean discordance 1.7 mm). Evaluators demonstrated greater accuracy in identifying the landmark in juveniles (1.4 mm) vs. adults (1.8 mm; P < 0.01). Results of this study indicated that observer repeatability is good and accuracy is fair for correctly identifying the radiographic cranial margin of the sacral lamina in dogs. This should be taken into consideration when interpreting elongation of the sacral lamina in radiographs of dogs with suspected DLSS, especially adults.     

THE EFFECT OF A TECHNICAL QUALITY ASSESSMENT OF HIP‐EXTENDED RADIOGRAPHS ON INTEROBSERVER AGREEMENT IN THE DIAGNOSIS OF CANINE HIP DYSPLASIA

Experienced and inexperienced observers evaluated the assessability of 50 radiographs (25 dogs) and determined the hip status (dysplasia/nondysplasia and final scoring according Fédération Cynologique Internationale [FCI]‐criteria) individually. A radiographic technical quality assessment was performed in a separate reading session. Interobserver agreement in determining dysplasia/nondysplasia and FCI‐scoring did not significantly increase with the increasing quality of a radiograph, irrespective whether these observers are experienced or not. There was a significant agreement between the technical quality assessment and assessability (P<0.0005). Despite the effort to objectify radiographic quality and to present high‐quality radiographs to observers, interobserver agreement on dysplasia/nondysplasia and final scoring, remains low, even in the experienced group. Although increased radiographic quality narrows the range of scoring, the range remains unacceptably high.     

VALIDATION OF A NOVEL TECHNIQUE FOR CREATING SIMULATED RADIOGRAPHS USING COMPUTED TOMOGRAPHY DATASETS

Understanding radiographic anatomy and the effects of varying patient and radiographic tube positioning on image quality can be a challenge for students. The purposes of this study were to develop and validate a novel technique for creating simulated radiographs using computed tomography (CT) datasets. A DICOM viewer (ORS Visual) plug‐in was developed with the ability to move and deform cuboidal volumetric CT datasets, and to produce images simulating the effects of tube‐patient‐detector distance and angulation. Computed tomographic datasets were acquired from two dogs, one cat, and one horse. Simulated radiographs of different body parts (n = 9) were produced using different angles to mimic conventional projections, before actual digital radiographs were obtained using the same projections. These studies (n = 18) were then submitted to 10 board‐certified radiologists who were asked to score visualization of anatomical landmarks, depiction of patient positioning, realism of distortion/magnification, and image quality. No significant differences between simulated and actual radiographs were found for anatomic structure visualization and patient positioning in the majority of body parts. For the assessment of radiographic realism, no significant differences were found between simulated and digital radiographs for canine pelvis, equine tarsus, and feline abdomen body parts. Overall, image quality and contrast resolution of simulated radiographs were considered satisfactory. Findings from the current study indicated that radiographs simulated using this new technique are comparable to actual digital radiographs. Further studies are needed to apply this technique in developing interactive tools for teaching radiographic anatomy and the effects of varying patient and tube positioning.     

OFF‐SITE SMARTPHONE VS. STANDARD WORKSTATION IN THE RADIOGRAPHIC DIAGNOSIS OF SMALL INTESTINAL MECHANICAL OBSTRUCTION IN DOGS AND CATS

Off‐site consultations by board‐certified veterinary radiologists benefit residents and emergency clinicians by providing immediate feedback and potentially improving patient outcome. Smartphone devices and compressed images transmitted by email or text greatly facilitate availability of these off‐site consultations. Criticism of a smartphone interface for off‐site consultation is mostly directed at image degradation relative to the standard radiographic viewing room and monitors. The purpose of this retrospective, cross‐sectional, methods comparison study was to compare the accuracy of abdominal radiographs in two imaging interfaces (Joint Photographic Experts Group, off‐site, smartphone vs. Digital Imaging and Communications in Medicine, on‐site, standard workstation) for the diagnosis of small intestinal mechanical obstruction in vomiting dogs and cats. Two board‐certified radiologists graded randomized abdominal radiographs using a five‐point Likert scale for the presence of mechanical obstruction in 100 dogs or cats presenting for vomiting. The area under the receiver operator characteristic curves for both imaging interfaces was high. The accuracy of the smartphone and traditional workstation was not statistically significantly different for either reviewer (P = 0.384 and P = 0.536). Correlation coefficients were 0.821 and 0.705 for each reviewer when the same radiographic study was viewed in different formats. Accuracy differences between radiologists were potentially related to years of experience. We conclude that off‐site expert consultation with a smartphone provides an acceptable interface for accurate diagnosis of small intestinal mechanical obstruction in dogs and cat.     

Radiographic Calibration for Analysis of Bone Mineral Density of the Equine Third Metacarpal Bone

Digital radiography represents the primary diagnostic tool the veterinarian uses to diagnose skeletal injuries in the horse. Advances in digital radiography have provided the veterinarian with opportunities to make simple radiographic assessments from calibrated digital radiographs such as dimensional analyses; however, more complex variables such as radiographic opacity have yet to be standardized. Therefore, we investigated the quantification of bone mineral density (BMD) via computed radiographic absorptiometry at various radiographic exposure intensities (kV), times (sec), and milliamps (mA) in the third metacarpal in the horse. By developing a brightness/darkness index (BDI), the grayscale of radiographs, calibrated with an aluminum (Al) marker of various known thicknesses and uniform densities, can be compared to the average BMD of a region of interest at various radiographic exposures. Al BDI was a significant predictor of bone BDI (r² = 0.960, P < .001) and BMD (r² = 0.971, P < .001). This method of calibration can be used for quantitative noninvasive bone mineral analysis and allows direct comparison of radiographs taken under different exposure settings.     

COMPARISON OF FLAT-PANEL DIGITAL TO CONVENTIONAL FILM-SCREEN RADIOGRAPHY IN DETECTION OF EXPERIMENTALLY CREATED LESIONS OF THE EQUINE THIRD METACARPAL BONE

Radiographic diagnosis of equine bone disease using digital radiography is prevalent in veterinary practice. However, the diagnostic quality of digital vs. conventional radiography has not been compared systematically. We hypothesized that digital radiography would be superior to film-screen radiography for detection of subtle lesions of the equine third metacarpal bone. Twenty-four third metacarpal bones were collected from horses euthanized for reasons other than orthopedic disease. Bones were dissected free of soft tissue and computed tomography was performed to ensure that no osseous abnormalities were present. Subtle osseous lesions were produced in the dorsal cortex of the third metacarpal bones, and the bones were radiographed in a soft tissue phantom using indirect digital and conventional radiography at standard exposures. Digital radiographs were printed onto film. Three Diplomates of the American College of Veterinary Radiology evaluated the radiographs for the presence or absence of a lesion. Receiver operator characteristic curves were constructed, and the area under these curves were compared to assess the ability of the digital and film-screen radiographic systems to detect lesions. The area under the ROC curves for film-screen and digital radiography were 0.87 and 0.90, respectively ( P =0.59). We concluded that the digital radiographic system was comparable to the film-screen system for detection of subtle lesions of the equine third metacarpal bone.     

ANATOMIC REFERENCE FOR COMPUTED TOMOGRAPHY OF THE HEAD OF THE FOAL

The purpose of this study was to produce an anatomic reference for computed tomography (CT) of the head of the foal for use by radiologists, clinicians, and veterinary students. The head from each of 2 foals, euthanized for reasons unrelated to head pathology, was removed and prepared for CT scanning. Using a third-generation CT scanner, 5-mm contiguous transverse images were acquired. The heads were then frozen and sectioned using a band saw, with the cuts matched as closely as possible to the CT slices. The anatomic sections were photographed and radiographed. The radiographs and anatomic photographs were digitized and matched with the corresponding CT image. Each CT image was compared with its corresponding radiographic and anatomic section to assist in the accurate identification of specific structures. Clinically relevant structures were identified and labeled in corresponding images (CT, anatomic slice, and radiograph of slice). Only structures identified in the CT image were labeled in 1 of the other 2 images. Sagittal (reference) images of the horse's head were reconstructed from the transverse CT scans, and were used to indicate the level from which each of the transverse images was obtained. Corresponding labeled images were then formatted together with a legend for identification of specific anatomic structures.     

A radiographic stress technique for evaluation of coxofemoral joint laxity in dogs

OBJECTIVE: To develop a radiographic stress technique to quantify hip joint laxity in dogs. STUDY DESIGN: Prospective study on client-owned dogs presented for hip dysplasia screening. ANIMAL POPULATION: 302 nonselected dogs (63 breeds). METHODS: Dogs were sedated and placed in dorsal recumbency. During pelvic radiography, the femoral heads were displaced manually in a craniodorsal direction. On these stress radiographs, the degree of lateral displacement of the femoral heads was assessed in terms of a subluxation index (SI) and compared with the degree of femoral head subluxation and the severity of hip dysplasia found on conventional extended hindlimb radiographs. RESULTS: The degree of subluxation on stress radiographs was significantly greater than on conventional radiographs. Correlation between the severity of canine hip dysplasia (CHD), graded according to conventional techniques, and the degree of subluxation, induced by radiographic stress technique, was positive (r = .57; P<.0001). Two critical SI values were noted. Of the dogs with an SI value of 0.3 or less, 99% were classified as CHD grade normal, borderline, or mildly dysplastic. In dogs with an SI value between 0.3 and 0.5, CHD grading ranged from normal to severely dysplastic. Of the dogs with an SI value greater than 0.5, 95% had dysplastic hip joints. CONCLUSIONS: Hip joint laxity cannot be quantified precisely on standard position radiographs. The proposed stress technique yields significantly higher degrees of femoral head subluxation than the standard position.     

PREVALENCE OF CANINE HIP DYSPLASIA IN A VETERINARY TEACHING HOSPITAL POPULATION

The purpose of this study was to determine the prevalence of canine HD in a population in which there was minimal or no prior screening of radiographs for the disorder. Patient information was obtained from the radiographic database at the University of Missouri-Columbia Veterinary Teaching Hospital during the five-year period of 1991-1995. The coxofemoral joints on ventrodorsal radiographs of the pelvis were independently evaluated by three veterinary radiologists. A consensus evaluation of normal, borderline, or dysplastic was compiled. There were 2885 dogs identified representing 116 breeds and the mixbreds. There were 2236 purebred dogs (1071 males and 1165 females) and the prevalence of HD was 19.7%. There were 649 mixbred dogs (340 males and 309 females) and the prevalence of HD was 17.7%. There was no significant difference in the prevalence of HD between sexes or between purebred and mixbred dogs (P = 0.16; P = 0.29). Degenerative joint disease (DJD) was the most common radiographic manifestation of HD and there appeared to be a threshold at 12 months of age after which the presence of DJD was the primary diagnostic criteria.