Deep learning-based detection of primary bone tumors around the knee joint on radiographs: a multicenter study.

Background: Most primary bone tumors are often found in the bone around the knee joint. However, the detection of primary bone tumors on radiographs can be challenging for the inexperienced or junior radiologist. This study aimed to develop a deep learning (DL) model for the detection of primary bone tumors around the knee joint on radiographs. Methods: From four tertiary referral centers, we recruited 687 patients diagnosed with bone tumors (including osteosarcoma, chondrosarcoma, giant cell tumor of bone, bone cyst, enchondroma, fibrous dysplasia, etc.; 417 males, 270 females; mean age 22.8±13.2 years) by postoperative pathology or clinical imaging/follow-up, and 1,988 participants with normal bone radiographs (1,152 males, 836 females; mean age 27.9±12.2 years). The dataset was split into a training set for model development, an internal independent and an external test set for model validation. The trained model located bone tumor lesions and then detected tumor patients. Receiver operating characteristic curves and Cohen's kappa coefficient were used for evaluating detection performance. We compared the model's detection performance with that of two junior radiologists in the internal test set using permutation tests. Results: The DL model correctly localized 94.5% and 92.9% bone tumors on radiographs in the internal and external test set, respectively. An accuracy of 0.964/0.920, and an area under the receiver operating characteristic curve (AUC) of 0.981/0.990 in DL detection of bone tumor patients were for the internal and external test set, respectively. Cohen's kappa coefficient of the model in the internal test set was significantly higher than that of the two junior radiologists with 4 and 3 years of experience in musculoskeletal radiology (Model vs. Reader A, 0.927 vs. 0.777, P<0.001; Model vs. Reader B, 0.927 vs. 0.841, P=0.033). Conclusions: The DL model achieved good performance in detecting primary bone tumors around the knee joint. This model had better performance than those of junior radiologists, indicating the potential for the detection of bone tumors on radiographs.

Accurate and Feasible Deep Learning Based Semi-Automatic Segmentation in CT for Radiomics Analysis in Pancreatic Neuroendocrine Neoplasms.

Current clinical practice or radiomics studies of pancreatic neuroendocrine neoplasms (pNENs) require manual delineation of the lesions in computed tomography (CT) images, which is time-consuming and subjective. We used a semi-automatic deep learning (DL) method for segmentation of pNENs and verified its feasibility in radiomics analysis. This retrospective study included two datasets: Dataset 1, contrast-enhanced CT images (CECT) of 80 and 18 patients respectively collected from two centers; and Dataset 2, CECT of 56 and 16 patients respectively from two centers. A DL-based semi-automatic segmentation model was developed and validated with Dataset 1 and Dataset 2, and the segmentation results were used for radiomics analysis from which the performance was compared against that based on manual segmentation. The mean Dice similarity coefficient of the trained segmentation model was 81.8% and 74.8% for external validation with Dataset 1 and Dataset 2 respectively. Four classifiers frequently used in radiomics studies were trained and tested with leave-one-out cross-validation strategy. For pathological grading prediction with Dataset 1, the area under the receiver operating characteristic curve (AUC) with semi-automatic segmentation was up to 0.76 and 0.87 respectively for internal and external validation. For recurrence study with Dataset 2, the AUC with semi-automatic segmentation was up to 0.78. All these AUCs were not statistically significant from the corresponding results based on manual segmentation. Our study showed that DL-based semi-automatic segmentation is accurate and feasible for the radiomics analysis in pNENs.

Improved shear wave motion detection using coded excitation for transient elastography.

Transient elastography (TE) is well adapted for use in studying liver elasticity. However, because the shear wave motion signal is extracted from the ultrasound signal, the weak ultrasound signal can significantly deteriorate the shear wave motion tracking process and make it challenging to detect the shear wave motion in a severe noise environment, such as within deep tissues and within obese patients. This paper, therefore, investigated the feasibility of implementing coded excitation in TE for shear wave detection, with the hypothesis that coded ultrasound signals can provide robustness to weak ultrasound signals compared with traditional short pulse. The Barker 7, Barker 13, and short pulse were used for detecting the shear wave in the TE application. Two phantom experiments and one in vitro liver experiment were done to explore the performances of the coded excitation in TE measurement. The results show that both coded pulses outperform the short pulse by providing superior shear wave signal-to-noise ratios (SNR), robust shear wave speed measurement, and higher penetration intensity. In conclusion, this study proved the feasibility of applying coded excitation in shear wave detection for TE application. The proposed method has the potential to facilitate robust shear elasticity measurements of tissue.

Preliminary study on the effect of stiffness on lamb wave propagation in bovine corneas.

The viscoelastic properties of human cornea could provide valuable information for various clinical applications. Particularly, it will be helpful to achieve a patient-specific biomechanical optimization in LASIK refractive surgery, early detection of corneal ecstatic disease or improved accuracy of intraocular pressure (IOP) measurement. However, there are few techniques that are capable of accurately assessing the corneal elasticity in situ in a nondestructive fashion. In order to develop a quantitative method for assessing both elasticity and viscosity of the cornea, we use ultrasound radiation force to excite Lamb waves in cornea, and a pulse echo transducer to track the tissue vibration. The fresh postmortem bovine eyes were treated via collagen cross-linking to make the cornea stiff. The effect of stiffness was studied by comparing the propagation of Lamb waves in normal and treated corneas. It was found that the waveform of generated Lamb waves changed significantly due to the increase in higher modes in treated corneas. This result indicated that the generated waveform was a complex of multiple harmonics and the varied stiffness will affect the energy distribution over different components. Therefore, it is important for assessing the viscoelastic properties of the cornea to know the components of Lamb wave and calculate the phase velocity appropriately.

Highly sensitive computer aided diagnosis system for breast tumor based on color Doppler flow images.

A computer-aided diagnosis (CAD) system for breast tumor based on color Doppler flow images is proposed. Our system consists of automatic segmentation, feature extraction, and classification of breast tumors. First, the B-mode grayscale image containing anatomical information was separated from a color Doppler flow image (CDFI). Second, the boundary of the breast tumor was automatically defined in the B-mode image and then morphologic and gray features were extracted. Third, an optimal feature vector was created using K-means cluster algorithm. Then a back-propagation (BP) artificial neural network (ANN) was used to classify breast tumors as benign, malignant or uncertain. Finally, the blood flow feature was extracted selectively from the CDFI, and was used to classify the uncertain tumor as benign or malignant. Experiments on 500 cases show that the proposed system yields an accuracy of 100% for the malignant and 80.8% for the benign classification. Comparing with other systems, the advantage of our system is that it has a much lower percentage of malignant tumor misdiagnosis.