Use of an artificial neural network to determine the diagnostic value of specific clinical and radiologic parameters in the diagnosis of interstitial lung disease on chest radiographs

Two-dimensional MR Spectroscopic Characterization of Breast Cancer In Vivo

The major goal of this work was to characterize invasive ductal carcinoma and healthy fatty breast tissues noninvasively using the classification and regression tree analysis (CART) of 2D MR spectral data. 2D L-COSY spectra were acquired in 14 invasive breast carcinoma and 21 healthy fatty breasts using a GE 1.5 Tesla MRI/MRS scanner equipped with a 2-channel phased-array breast MR coil. The 2D spectra were recorded in approximately 10 minutes using a minimum voxel size of 1 ml without any water suppression technique. For healthy breasts, spectra were acquired from at least one fatty region. 2D L-COSY spectra were recorded in a total of 43 voxels. Five diagonal and six cross peak volumes were integrated and at least eighteen ratios were selected as potential features for the statistical method, namely CART. The 2D L-COSY data showed a significant increase for the majority of these ratios in invasive breast carcinomas compared to healthy fatty tissues. Better accuracy of identifying carcinomas and fatty tissues is reported using CART analysis of different combinations of ratios calculated from the relative levels of water, choline, and saturated and unsaturated lipids. This is a first report on the statistical classification of 2D L-COSY in human breast carcinomas in vivo.

A New Approach to Improve TMJ Morphological Information from Plain Film Radiographs

The hard tissues of the temporomandibular joint (TMJ) can be assessed through radiographic imaging to provide information to assist in diagnosis and treatment. However, the quality of information gathered from such imaging is often less than desired due to the small size of the TMJ, the widely varying fossa and condylar morphology, and the surrounding dense osseous structures. These make a clear and undistorted radiographic image of the hard tissue of the joint technically difficult. It is postulated that, if the degree of inaccuracy of a given radiograph is known quantitatively and taken into account, the clinician will be able to make a better informed qualitative assessment of TMJ morphology. The aim of this study is: 1. to improve the qualitative information that can be acquired from routine clinical plain film radiographs of the TMJ; 2. to use quantitative data to test the novel Neural Network (NN) model; and 3. to statistically show the discrepancies between routine radiographic images and the actual joint. Linear measurements of excised TMJs and their radiographic images were used to train a multilayer perceptron (MP) type NN model to predict joint dimensions more accurately. Such a neural network, developed using a statistical software package such as SPSS (SPSS, Inc. Chicago, IL), functions as a computer software program and predicts joint dimensions with increased accuracy when radiographic measurements are entered into the program. The NN model described here predicts the actual linear distances of the TMJ more closely than radiographic measurements and is more reliable in assessing the TMJ morphology.

Computer-assisted detection of infectious lung diseases: a review

Respiratory tract infections are a leading cause of death and disability worldwide. Although radiology serves as a primary diagnostic method for assessing respiratory tract infections, visual analysis of chest radiographs and computed tomography (CT) scans is restricted by low specificity for causal infectious organisms and a limited capacity to assess severity and predict patient outcomes. These limitations suggest that computer-assisted detection (CAD) could make a valuable contribution to the management of respiratory tract infections by assisting in the early recognition of pulmonary parenchymal lesions, providing quantitative measures of disease severity and assessing the response to therapy. In this paper, we review the most common radiographic and CT features of respiratory tract infections, discuss the challenges of defining and measuring these disorders with CAD, and propose some strategies to address these challenges.

Neural network evaluation of PET scans of the liver: a potentially useful adjunct in clinical interpretation

PURPOSE

To assess the performance of an artificial neural network in the evaluation of fluorine 18 fluorodeoxyglucose (FDG) uptake in the liver, compared with the results of expert interpretation of abdominal liver magnetic resonance (MR) images.

MATERIALS AND METHODS

The study was approved by the institutional human research committee and was HIPAA compliant, with waiver of informed consent. Digital data from positron emission tomographic (PET)/computed tomographic (CT) examinations, along with patient demographics, were obtained from 98 consecutive patients who underwent both whole-body PET/CT examinations and liver MR imaging examinations within 2 months. Interpretations of the scans from PET/CT examinations by trained neural networks were cross-classified with expert interpretations of the findings on images from MR examinations for intrahepatic benignity or malignancy. Receiver operating characteristic (ROC) curves were obtained for the designed networks. The significance of the difference between neural network ROC curves and the ROC curves detailing the performance of two expert blinded observers in the interpretation of liver FDG uptake was determined.

RESULTS

A neural network incorporating lesion data demonstrated an ROC curve with an area under the curve (AUC) of 0.905 (standard error, 0.0370). A network independent of lesion data demonstrated an ROC curve with an AUC of 0.896 (standard error, 0.0386). These results compare favorably with results of expert blinded observers 1 and 2 who demonstrated ROCs with AUCs of 0.786 (standard error, 0.0522) and 0.796 (standard error, 0.0514), respectively. Following unblinding to network data, the AUCs for readers 1 and 2 improved to 0.924 (standard error, 0.0335) and 0.881 (standard error, 0.0409), respectively.

CONCLUSION

Computers running artificial neural networks employing PET/CT scan data are sensitive and specific in the designation of the presence of intrahepatic malignancy, with comparison with interpretation by expert observers. When used in conjunction with human expertise, network data improve accuracy of the human interpreter.

Differentiation of common large sellar-suprasellar masses effect of artificial neural network on radiologists' diagnosis performance

RATIONALE AND OBJECTIVES

When pituitary adenoma, craniopharyngioma, and Rathke's cleft cyst grow in the sellar and suprasellar region, it is often difficult to differentiate among these three lesions on magnetic resonance (MR) images. The purpose of this study was to apply an artificial neural network (ANN) for differential diagnosis among these three lesions with MR images and retrospectively evaluate the effect of ANN output on radiologists' performance.

MATERIALS AND METHODS

Forty-three patients with sellar-suprasellar masses were studied. The ANN was designed to differentiate among pituitary adenoma, craniopharyngioma, and Rathke's cleft cyst by using patients' ages and nine MR image findings obtained by three neuroradiologists using a subjective rating scale. In the observer performance test, MR images were viewed by nine radiologists, including four neuroradiologists and five general radiologists, first without and then with ANN output. The radiologists' performance was evaluated using receiver-operating characteristic analysis with a continuous rating scale.

RESULTS

The ANN showed high performance in differentiation among the three lesions (area under the receiver-operating characteristic curve, 0.990). The average area under the curve for all radiologists for differentiation among the three lesions increased significantly from 0.910 to 0.985 (P = .0024) when they used the computer output. Areas under the curves for the general radiologists and neuroradiologists increased from 0.876 to 0.983 (P = .0083) and from 0.952 to 0.989 (P = .038), respectively.

CONCLUSION

In diagnostic performance for differentiation among pituitary macroadenoma, craniopharyngioma, and Rathke's cleft cyst with MR imaging, the ANN resulted in parity between neuroradiologists and general radiologists.

Investigation of breast cancer using two-dimensional MRS

Proton (1H) MRS enables non-invasive biochemical assay with the potential to characterize malignant, benign and healthy breast tissues. In vitro studies using perchloric acid extracts and ex vivo magic angle spinning spectroscopy of intact biopsy tissues have been used to identify detectable metabolic alterations in breast cancer. The challenges of 1H MRS in vivo include low sensitivity and significant overlap of resonances due to limited chemical shift dispersion and significant inhomogeneous broadening at most clinical magnetic field strengths. Improvement in spectral resolution can be achieved in vivo and in vitro by recording the MR spectra spread over more than one dimension, thus facilitating unambiguous assignment of metabolite and lipid resonances in breast cancer. This article reviews the recent progress with two-dimensional MRS of breast cancer in vitro, ex vivo and in vivo. The discussion includes unambiguous detection of saturated and unsaturated fatty acids, as well as choline-containing groups such as free choline, phosphocholine, glycerophosphocholine and ethanolamines using two-dimensional MRS. In addition, characterization of invasive ductal carcinomas and healthy fatty/glandular breast tissues non-invasively using the classification and regression tree (CART) analysis of two-dimensional MRS data is reviewed.

Computer-aided diagnosis for the differentiation of malignant from benign thyroid nodules on ultrasonography

RATIONALE AND OBJECTIVES

We sought to evaluate the diagnostic performance of an artificial neural network (ANN) and binary logistic regression (BLR) in differentiating malignant from benign thyroid nodules on ultrasonography.

MATERIALS AND METHODS

Two experienced radiologists, who were unaware of the histopathological diagnosis, analyzed ultrasonographic (US) features of 109 pathologically proven thyroid lesions (49 malignant and 60 benign) in 96 patients. Each radiologist was asked to evaluate US findings and categorize nodules into one of the two groups (malignant vs. benign) in each case. The following 8 US parameters were assessed for each nodule: size, shape, margin, echogenicity, cystic change, microcalcification, macrocalcification, and halo sign. Statistically significant US findings were obtained with backward stepwise logistic regression and were used for training and testing of the ANN and the BLR. The performance of the ANN and BLR was compared to that of the radiologists using receiver-operating characteristic (ROC) analysis.

RESULTS

Statistically significant US findings were size, margin, echogenicity, cystic change, and macrocalcification of the nodules. The area under the ROC curve (Az) values of ANN and BLR were 0.9492 +/- 0.0195 and 0.9046 +/- 0.0289, respectively. The Az value was 0.8300 +/- 0.0359 for reader 1 and 0.7600 +/- 0.0409 for reader 2. The Az values for ANN and BLR were significantly higher than those for both radiologists (all p < .05).

CONCLUSION

The performance of the ANN and the BLR was better than that of the radiologists in the distinction of benign and malignant thyroid nodules.

Improvement of lung abnormality detection in computed radiography using multi-objective frequency processing: Evaluation by receiver operating characteristics (ROC) analysis

Computed radiography (CR) has been shown to have relatively low sensitivity for detection of pulmonary nodules. This poor sensitivity precludes its use as a screening modality despite the low cost, low dose and wide distribution of devices. The purpose of this study was to apply multi-objective frequency processing (MFP) to CR images and to evaluate its usefulness for diagnosing subtle lung abnormalities. Fifty CR images with simulated subtle lung abnormalities were obtained from 50 volunteers. Each image was processed with MFP. We cut chest images. The chest image was divided into two rights and left. A total of 200 half-chest images (100 MFP-processed images and 100 MFP-unprocessed images) were prepared. Five radiologists participated in this study. ROC analyses demonstrated that the detection rate of simulated subtle lung abnormalities on the CR images was significantly better with MFP (Az=0.8508) than without MFP (Az=0.7925). The CR images processed with MFP could be useful for diagnosing subtle lung abnormalities. In conclusion, MFP appears to be useful for increasing the sensitivity and specificity in the detection of pulmonary nodules, ground-glass opacity (GGO) and reticular shadow.

Computer-aided diagnosis in chest radiography

We have developed computer-aided diagnosis (CAD) schemes for the detection of lung nodules, interstitial lung diseases, interval changes, and asymmetric opacities, and also for the differential diagnosis of lung nodules and interstitial lung diseases on chest radiographs. Observer performance studies indicate clearly that radiologists' diagnostic accuracy was improved significantly when radiologists used a computer output in their interpretations of chest radiographs. In addition, the automated recognition methods for the patient and the projection view by use of chest radiographs were useful for integrating the chest CAD schemes into the picture-archiving and communication system (PACS).

Usefulness of artificial neural network for differential diagnosis of hepatic masses on CT images

RATIONALE AND OBJECTIVE

Our purpose in this study is to apply an artificial neural network (ANN) for differential diagnosis of certain hepatic masses on computed tomographic (CT) images and evaluate the effect of ANN output on radiologist diagnostic performance.

MATERIALS AND METHODS

We collected 120 cases of hepatic disease. We used a single three-layer feed-forward ANN with a back-propagation algorithm. The ANN is designed to differentiate four hepatic masses (hepatocellular carcinoma, intrahepatic peripheral cholangiocarcinoma, hemangioma, and metastasis) by using nine clinical parameters and 24 radiological findings in dual-phase contrast-enhanced CT images. Thus, the ANN consisted of 33 input units and four output units. Subjective ratings for the 24 radiological findings were provided independently by two attending radiologists. All clinical cases were used for training and testing of the ANN by implementation of a round-robin technique. In the observer test, CT images of all 120 cases (30 cases for each disease) were used. CT images were viewed by seven radiologists first without and then with ANN output. Radiologist performance was evaluated by using receiver operating characteristic (ROC) analysis on a continuous rating scale.

RESULTS

Averaged area under the ROC curve for ANN alone was 0.961. The diagnostic performance of seven radiologists increased from 0.888 to 0.934 (P < .02) when they used ANN output.

CONCLUSION

The ANN can provide useful output as a second opinion to improve radiologist diagnostic performance in the differential diagnosis of hepatic masses seen on contrast-enhanced CT.

Prediction of burn healing time using artificial neural networks and reflectance spectrometer

BACKGROUND

Burn depth assessment is important as early excision and grafting is the treatment of choice for deep dermal burn. Inaccurate assessment causes prolonged hospital stay, increased medical expenses and morbidity. Based on reflected burn spectra, we have developed an artificial neural network to predict the burn healing time.

PURPOSE

Our study is to develop a non-invasive objective method to predict burn-healing time.

METHODS AND MATERIALS

Burns less than 20% TBSA was included. Burn spectra taken on the third postburn day using reflectance spectrometer were analyzed by an artificial neural network system.

RESULTS

Forty-one spectra were collected. With the newly developed method, the predictive accuracy of burns healed in less than 14 days was 96%, and that in more than 14 days was 75%.

CONCLUSIONS

Using reflectance spectrometer, we have developed an artificial neural network to determine the burn healing time with 86% overall predictive accuracy.

Neural network approach to the segmentation and classification of dynamic magnetic resonance images of the breast: comparison with empiric and quantitative kinetic parameters

RATIONALE AND OBJECTIVE

An artificial neural network (ANN)-based segmentation method was developed for dynamic contrast-enhanced magnetic resonance (MR) imaging of the breast and compared with quantitative and empiric parameter mapping techniques.

MATERIALS AND METHODS

The study population was composed of 10 patients with seven malignant and three benign lesions undergoing dynamic MR imaging of the breast. All lesions were biopsied or surgically excised, and examined by means of histopathology. A T1-weighted 3D FLASH (fast low angle shot sequence) was acquired before and seven times after the intravenous administration of gadopentetate dimeglumine at a dose of 0.1 mmol/kg body weight. Motion artifacts on MR images were eliminated by voxel-based affine and nonrigid registration techniques. A two-layered feed-forward back-propagation network was created for pixel-by-pixel classification of signal intensity-time curves into benign/malignant tissue types. ANN output was statistically compared with percent-enhancement (E), signal enhancement ratio (SER), time-to-peak, subtracted signal intensity (SUB), pharmacokinetic parameter rate constant (k(ep)), and correlation coefficient to a predefined reference washout curve.

RESULTS

ANN was successfully applied to the classification of breast MR images identifying structures with benign or malignant enhancement kinetics. Correlation coefficient (logistic regression, odds ratio [OR] = 12.9; 95% CI: 7.7-21.8), k(ep) (OR = 1.8; 95% CI: 1.2-2.6), and time-to-peak (OR = 0.45; 95% CI: 0.3-0.7) were independently associated to ANN output classes. SER, E, and SUB were nonsignificant covariates.

CONCLUSION

ANN is capable of classifying breast lesions on MR images. Mapping correlation coefficient, k(ep) and time-to-peak showed the highest association with the ANN result.

Application of an Artificial Neural Network to High-Resolution CT: Usefulness in Differential Diagnosis of Diffuse Lung Disease

OBJECTIVE

The purpose of our study was to evaluate the diagnostic performance of an artificial neural network (ANN) in differentiating among certain diffuse lung diseases using high-resolution CT (HRCT) and the effect of ANN output on radiologists' diagnostic performance.

MATERIALS AND METHODS

We selected 130 clinical cases of diffuse lung disease. We used a single three-layer, feed-forward ANN with a back-propagation algorithm. The ANN was designed to differentiate among 11 diffuse lung diseases by using 10 clinical parameters and 23 HRCT features. Therefore, the ANN consisted of 33 input units and 11 output units. Subjective ratings for 23 HRCT features were provided independently by eight radiologists. All clinical cases were used for training and testing of the ANN by implementing a round-robin technique. In the observer test, a subset of 45 cases was selected from the database of 130 cases. HRCT images were viewed by eight radiologists first without and then with ANN output. The radiologists' performance was evaluated with receiver operating characteristic (ROC) analysis with a continuous rating scale.

RESULTS

The average area under the ROC curve for ANN performance obtained with all clinical parameters and HRCT features was 0.956. The diagnostic performance of four chest radiologists and four general radiologists was increased from 0.986 to 0.992 (p = 0.071) and 0.958 and 0.971 (p < 0.001), respectively, when they used the ANN output based on their own feature ratings.

CONCLUSION

The ANN can provide a useful output as a second opinion to improve general radiologists' diagnostic performance in the differential diagnosis of certain diffuse lung diseases using HRCT.

Application of artificial neural networks to the analysis of dynamic MR imaging features of the breast

The discriminative ability of established diagnostic criteria for MRI of the breast is assessed, and their relative relevance using artificial neural networks (ANNs) is determined. A total of 89 women with 105 histopathologically verified breast lesions (73 invasive cancers, 2 in situ cancers, and 30 benign lesions) were included in this study. A T1-weighted 3D FLASH sequence was acquired before and seven times after the intravenous administration of gadopentetate dimeglumine at a dose of 0.2 mmol/kg body weight. ANN models were built to test the discriminative ability of kinetic, morphologic, and combined MR features. The subjects were randomly divided into two parts: a training set of 59 lesions and a verification set of 46 lesions. The training set was used for learning, and the performance of each model was evaluated on the verification set by measuring the area under the ROC curve (Az). An optimally minimized model was constructed using the most relevant input variables that were determined by the automatic relevance determination (ARD) method. ANN models were compared with the performance of a human reader. Margin type, time-to-peak enhancement, and washout ratio showed the highest discriminative ability among diagnostic criteria and comprised the minimized model. Compared with the expert radiologist (Az = 0.799), using the same prediction scale, the minimized ANN model performed best (Az = 0.771), followed by the best kinetic (Az = 0.743), the maximized (Az = 0.727), and the morphologic model (Az = 0.678). The performance of a neural network prediction model is comparable to that of an expert radiologist. A neurostatistical approach is preferred for the analysis of diagnostic criteria when many parameters are involved and complex nonlinear relationships exist in the data set.

Artificial neural networks (ANNs) for differential diagnosis of interstitial lung disease : results of a simulation test with actual clinical cases1

RATIONALE AND OBJECTIVES

To evaluate the performance of an artificial neural network (ANN) scheme with use of consecutive clinical cases and its effect on radiologists with an observer test.

MATERIALS AND METHODS

Artificial neural networks were designed to distinguish among 11 interstitial lung diseases on the basis of 26 inputs (16 radiologic findings, 10 clinical parameters). Chest radiographs of 96 consecutive cases with interstitial lung disease were used. Five radiologists independently rated their radiologic findings on the 96 chest radiographs. Based on their ratings of radiologic findings and clinical parameters obtained from the hospital information system, the output values indicating the likelihood of each of the 11 interstitial lung diseases were determined. Subsequently, 30 cases were selected from these 96 cases for an observer test. Five radiologists marked their confidence levels for diagnosis of 11 possible diseases in each case without and with ANN output. The performance of ANNs and radiologists was evaluated by receiver operating characteristic analysis based on their outputs and on confidence levels, respectively. RESULTS; The average Az value (area under the receiver operating characteristic curve) indicating ANN performance for the 96 consecutive cases was 0.85 +/- 0.03. The average Az values indicating radiologists' performance without and with ANN outputs were 0.81 +/- 0.11 and 0.87 +/- 0.06, respectively. The diagnostic accuracy was improved significantly when radiologists read chest radiographs with ANN outputs (P < .05).

CONCLUSION

Artificial neural networks for differential diagnosis of interstitial lung disease may be useful in clinical situations, and radiologists may be able to utilize the ANN output to their advantage in the differential diagnosis of interstitial lung disease on chest radiographs.

Computer-aided diagnosis in chest radiography: results of large-scale observer tests at the 1996-2001 RSNA scientific assemblies

Since 1996, computer-aided diagnosis (CAD) schemes have been presented as interactive demonstrations on computer workstations at each scientific assembly of the Radiological Society of North America. The schemes involved (a) detection of pulmonary nodules, (b) temporal subtraction, (c) detection of interstitial lung disease, (d) differential diagnosis of interstitial lung disease, and (e) distinction between benign and malignant pulmonary nodules on chest radiographs. Large-scale observer tests were carried out to examine how radiologists can benefit from CAD systems. Observer performance was evaluated by analysis of receiver operating characteristic (ROC) curves. The statistical significance of the difference between the areas under the ROC curves without and with CAD was analyzed with the Student t test. In all of the tests, the diagnostic accuracy of the radiologists in total improved significantly when CAD was used. This result provides additional evidence that CAD has the potential to improve the performance of radiologists in their decision-making process in interpreting chest radiographs.