H. MacMahon

IMAGE FEATURE ANALYSIS OF FALSE-POSITIVE DIAGNOSES PRODUCED BY AUTOMATED DETECTION OF LUNG NODULES

RATIONALE AND OBJECTIVES To reduce the number of false-negative diagnoses by radiologists, the authors are developing a computer-aided diagnosis scheme for detection of lung nodules in digital chest images. In this study, the authors attempted to reduce the number of false-positive diagnoses obtained with a previous computer scheme by incorporating additional knowledge from experienced chest radiologists into the computer scheme. METHODS The authors applied their previous computer scheme, using less-strict criteria, to 60 clinical chest radiographs; this yielded 735 candidate nodules (23 true nodules and 712 false-positive diagnoses). These candidates were analyzed using region-growing, trend-correction, and edge-gradient techniques to determine measures by which to quantify image features of candidate nodules. RESULTS The 712 false-positive diagnoses represented various anatomic structures that were located throughout the chest image. From this analysis, we were able to decrease the number of false-positive errors from an average of 12 to approximately 5 per image without eliminating any true nodules. CONCLUSION Our results show that incorporating knowledge from experienced chest radiologists into the computer algorithm will play an important role in the development of computerized schemes for the detection of pulmonary nodules.

Computer-aided diagnosis in chest radiography. Preliminary experience.

RATIONALE AND OBJECTIVES.Computer-aided diagnosis (CAD) schemes for chest radiography are being developed with which to alert radiologists to possible lesions, and thus potentially improve diagnostic accuracy. However, CAD schemes have not been tested on a large number of clinical cases. The authors identify design parameters that would be required for development of an intelligent workstation. METHODS.Computer-aided diagnosis programs were applied for the automated detection of lung nodules, cardiomegaly, and interstitial infiltrates to 310 consecutive chest radiographs, and were analyzed for potential usefulness and limitations. Computer-aided diagnosis output was evaluated by radiologists and physicists for accuracy and technical problems, respectively. RESULTS.Approximately 70% of the results were judged to be potentially acceptable; however, the number of false-positive findings was relatively high. Technical problems included failure to detect subtle abnormalities and the occurrence of false-positive detections caused by normal anatomical structures. CONCLUSION.Computer-aided diagnosis has the potential to be a valuable aid to radiologists in clinical practice, if certain technical problems can be overcome and if optimal operating points can be defined for clinical use.

Computerized Analysis of Lung Textures for Detection and Characterization of Interstitial Diseases in Chest Images

In order to detect and characterize interstitial disease, we are developing an automated method for determination of physical texture measures in digital chest radiographs. This method is based on an analysis of the power spectrum of lung texture. First, square ROIs are sampled from inter-rib spaces. The non-uniform background trend caused by the gross anatomy of the lung and chest wall is corrected to isolate the fluctuating patterns of the underlying lung texture. Finally, the rootsquare (rms) variation and the first moment of the power spectrum are determined as quantitative measures of the magnitude and coarseness (or fineness), respectively, of the lung texture. In addition, we devised an automated classification method for distinction normal and abnormal lungs with interstitial disease. A comparison of ROC curves obtained by radiologists and by means of the computerized method suggests that the computerized approadh may provide performance similar to human observers in distinguishing lungs with mild interstitial diseases from normal lungs. Moreover, texture measures obtained from this computer analysis of ILO pneumoconioses standard radiographs correspond closely with ILO classification categories. We believe tbat this technique has the potential to be a valuable aid to radiologists.

Automated Computerized Analysis Of Radiographic Images: An Intelligent Component Of IMAC To Aid Diagnosis

A weakness of the impact of the current IMAC and PACS on diagnostic radiology is partly due to the lack of an “intelligent” component which could assist radiologists in their diagnosis. An intelligent component of PACS can be implemented by automated computerized analysis of radiographic images. Computer output from quantitative analysis of digital images can be used as a “second opinion” to alert the radiologist by indicating potential lesion sites and/or by providing objective measurements of normal and abnormal patterns. The use of computer output in this manner is expected to improve diagnostic accuracy by reducing false negatives and U) improve the overall reproducibility of image interpretation. Current results of computerized automated analysis are demonstrated for the identification of lung nodules and pneumothoraces, and for the assessment of interstitial disease and cardiomegaly in chest radiography. These early results are very encouraging. Computer-aided diagnosis, which refers to a diagnosis made by a radiologist who utilizes the computer output of quantitative image analysis, may be a useful component of PACS which can provide practical benefits for radiologists.