One challenge in applying deep learning to medical imaging is the lack of labeled data. Although large amounts of clinical data are available, acquiring labeled image data is difficult, especially for bone scintigraphy (i.e., 2D bone imaging) images. Bone scintigraphy images are generally noisy, and ground-truth or gold standard information from surgical or pathological reports may not be available. We propose a novel neural network model that can segment abnormal hotspots and classify bone cancer metastases in the chest area in a semisupervised manner. Our proposed model, called MaligNet, is an instance segmentation model that incorporates ladder networks to harness both labeled and unlabeled data. Unlike deep learning segmentation models that classify each instance independently, MaligNet utilizes global information via an additional connection from the core network. To evaluate the performance of our model, we created a dataset for bone lesion instance segmentation using labeled and unlabeled example data from 544 and 9,280 patients, respectively. Our proposed model achieved mean precision, mean sensitivity, and mean F1-score of 0.852, 0.856, and 0.848, respectively, and outperformed the baseline mask region-based convolutional neural network (Mask R-CNN) by 3.92%. Further analysis showed that incorporating global information also helps the model classify specific instances that require information from other regions. On the metastasis classification task, our model achieves a sensitivity of 0.657 and a specificity of 0.857, demonstrating its great potential for automated diagnosis using bone scintigraphy in clinical practice.

Pendred syndrome is an autosomal recessive disorder characterized by congenital sensorineural deafness, goiter, and impaired iodide organification. It is caused by mutations in the PDS gene. Most published mutation studies of Pendred syndrome have dealt with Western populations. In this study, we examined clinical and molecular characteristics of 16 affected individuals in 6 unrelated Thai families. Of all the affected, 100% (16/16) had bilateral deafness, 68.8% (11/16) goiters, and 25% (4/16) hypothyroidism. Follicular thyroid carcinoma and Hürthle cell adenoma were found in affected members of a family, raising the possibility of an increased risk of thyroid carcinoma in Pendred syndrome patients. Sequence analysis of the entire coding region of the PDS gene successfully identified all 12 mutant alleles in these 6 families. The 12 identified mutant alleles constituted 6 distinct mutations including 3 splice site mutations (IVS4-1G>A, IVS7-2A>G, IVS9- 1G>A), one frame shift mutation (1548insC) and 2 missense mutations (T67S, H723R). Eight mutations out of 12 were constituted by IVS7- 2A>G and 1548insC, each one being present in 4 distinct alleles in our studied group. The identification of these two frequent PDS mutations will facilitate the molecular diagnosis of Pendred syndrome in Thai populations. In addition, three newly identified mutations, T67S, IVS4-1G>A, and IVS9-1G>A, were not observed in 50 unrelated healthy Thai controls.

Objectives: Myocardial perfusion scintigraphy (MPS) is an important diagnostic test for detecting of coronary artery stenosis (CAS); however, tissue attenuation can lead to a difference in accuracy. We evaluated the diagnostic accuracy of attenuation-corrected (AC) and non-attenuation-corrected (NC) MPS for the detection of CAS. Methods: We retrospectively recruited patients who underwent invasive coronary angiography within 10 months after Tc-99m sestamibi MPS. The AC and NC perfusion images were analyzed separately, and each myocardial segment was scored based on relative uptake from 0 to 4. The summed stress score (SSS), summed rest score (SRS), and summed difference score (SDS) were calculated. The diagnostic performances were analyzed using the area under the curve (AUC) of the receiver operating characteristic curve. Results: From 117 patients, significant coronary stenosis was present in 66 patients (56%). The SSS and SRS obtained from NC-images were higher than those from AC, supporting the presence of attenuation artifacts in NC images. The AUC of SSS and SDS were significantly higher than those of SRS in both AC- and NC-images, but no significant difference was found between the AUC of SSS, and those of SDS. The optimal cut-offs were >12 for AC-SSS, >15 for NC-SSS, >4 for AC-SDS and >3 for NC-SDS. There was no statistically significant difference in the sensitivity, specificity, positive predictive value, negative predictive value, and accuracy among AC-SSS, NC-SSS, AC-SDS, and NC-SDS. Conclusion: NC-based Tc-99m-sestamibi MPS promised comparable accuracy to AC images by using different cut-off values for diagnosis.

Background False negative myocardial perfusion images on single-photon emission computed tomography (SPECT) myocardial perfusion imaging (MPI) is a substantial problem in the interpretation of MPI. To explore possible parameters from gated SPECT that could detected coronary artery disease (CAD) in patients with normal perfusion images, we retrospectively selected patients who underwent a 2-day Tc-99 m MIBI adenosine stress/rest MPI and a subsequent invasive coronary angiography. Gated SPECT parameters, including end systolic volume (ESV), end diastolic volume (EDV), left ventricular ejection fraction (LVEF), and transient ischemic dilation (TID) ratio of patients with and without CAD were compared and assessed for their respective diagnostic performance using receiver operating characteristics (ROC) area under the curve (AUC). Results Among 22 patients with normal perfusion images, 7 patients had CAD. Despite the small number of patients, we found significant differences between the ESV, the EDV, and the LVEF between patients with and without CAD. The analysis of ROC curve showed the stress ESV could excellently detect CAD (AUC = 0.900). The rest ESV, the stress EDV, the rest EDV, the stress LVEF and the rest LVEF could likewise perform well in the detection of CAD (AUC = 0.833, 0.819, 0.790, 0.862, and 0.838, respectively). In contrast, the change in LVEF and the TID ratio (AUC = 0.667 and 0.524, respectively) did not seem as reliable as other parameters. Optimal cutoffs for detection of CAD in patients with normal perfusion images from our study were ≥ 20.0 mL for the stress ESV, ≥ 71.0 mL for the stress EDV, ≤ 66.3 EF units for the stress LVEF, ≥ 18.0 mL for the rest ESV, ≥ 67.0 mL for the rest EDV and ≤ 70.0 EF unit for the rest LVEF. Conclusions Gated SPECT parameters could facilitate detection of CAD in patient with normal perfusion images on Tc-99m MIBI MPI. These parameters should be carefully interpreted to improve diagnostic accuracy and reduce false negative MPI.