Objectives
To investigate machine learning approaches for radiomics-based prediction of prognostic biomarkers and molecular subtypes of breast cancer using quantification of tumor heterogeneity and ...angiogenesis properties on magnetic resonance imaging (MRI).
Methods
This prospective study examined 291 invasive cancers in 288 patients who underwent breast MRI at 3 T before treatment between May 2017 and July 2019. Texture and perfusion analyses were performed and a total of 160 parameters for each cancer were extracted. Relationships between MRI parameters and prognostic biomarkers were analyzed using five machine learning algorithms. Each model was built using only texture features, only perfusion features, or both. Model performance was compared using the area under the receiver-operating characteristic curve (AUC) and the DeLong method, and the importance of MRI parameters in prediction was derived.
Results
Texture parameters were associated with the status of hormone receptors, human epidermal growth factor receptor 2, and Ki67, tumor size, grade, and molecular subtypes (
p
< 0.002). Perfusion parameters were associated with the status of hormone receptors and Ki67, grade, and molecular subtypes (
p
< 0.003). The random forest model integrating texture and perfusion parameters showed the highest performance (AUC = 0.75). The performance of the random forest model was the best with a special scale filter of 0 (AUC = 0.80). The important parameters for prediction were texture irregularity (entropy) and relative extracellular extravascular space (
V
e
).
Conclusions
Radiomic machine learning that integrates tumor heterogeneity and angiogenesis properties on MRI has the potential to noninvasively predict prognostic factors of breast cancer.
Key Points
•
Machine learning, integrating tumor heterogeneity and angiogenesis properties on MRI, can be applied to predict prognostic biomarkers and molecular subtypes in breast cancer.
•
The random forest model showed the best predictive performance among the five machine learning models (logistic regression, decision tree, naïve Bayes, random forest, and artificial neural network).
•
The most important MRI parameters for predicting prognostic factors in breast cancer were texture irregularity (entropy) among texture parameters and relative extracellular extravascular space (V
e
) among perfusion parameters.
Our aim was to correlate chest CT and pathologic findings of polyhexamethylene guanidine phosphate (PHMG)-induced lung injuries in a rat model, to determine whether PHMG exposure causes lung tumors, ...and to explore genetic alterations according to PHMG exposure under the guidance of CT. A PHMG solution was intratracheally administrated to 40 male rats. Chest CT was carried out in all rats and both lungs were collected for histopathologic evaluation. At 4- and 8-weeks post-instillation, one lobe of the right lung from 3 rats was subjected to RNA sequencing. At least one abnormal CT finding was found in all rats at all weeks. The major CT findings were inflammation, fibrosis, and tumors in the pathologic analysis, where significant changes were observed over time. The lung lesions remained persistent after 8 weeks of PHMG exposure. In the pathologic analysis, the extent/severity of inflammation did not show statistically significant changes over time, whereas the extent/severity of fibrosis increased continuously up to 6 weeks after PHMG exposure and then decreased significantly at 8 weeks. Bronchiolar-alveolar adenomas which have malignant potential were found in 50% of rats at 6 and 8 weeks after PHMG exposure. Also, several genes associated with lung cancer, acute lung injury, and pulmonary fibrosis were detected. Our study revealed that PHMG-induced lung injury and its changes according to the number of weeks after exposure were demonstrated using chest CT and pathologic evaluation. In addition, we showed that PHMG exposure caused lung tumors and genetic alterations according to PHMG exposure under the guidance of CT.
We aimed to investigate the effect of chronic particulate matter (PM) exposure on bleomycin-induced lung fibrosis in a rat model using chest CT, histopathologic evaluation, and RNA-sequencing. A ...bleomycin solution was intratracheally administrated to 20 male rats. For chronic PM exposure, after four weeks of bleomycin treatment to induce lung fibrosis, PM suspension (experimental group) or normal saline (control group) was intratracheally administrated for 10 weeks. Chest CT was carried out in all rats, and then both lungs were extracted for histopathologic evaluation. One lobe from three rats in each group underwent RNA sequencing, and one lobe from five rats in each group was evaluated by western blotting. Inflammation and fibrosis scores in both chest CT and pathologic analysis were significantly more aggravated in rats with chronic PM exposure than in the control group. Several genes associated with inflammation and immunity were also upregulated with chronic PM exposure. Our study revealed that chronic PM exposure in a bleomycin-induced lung fibrosis rat model aggravated pulmonary fibrosis and inflammation, proven by chest CT, pathologic analysis, and RNA sequencing.
•Risk factors for kyphosis after laminoplasty in ossification of the posterior longitudinal ligament / cervical spondylotic myelopathy patients were evaluated.•A decrease in C2-7 lordosis and ...increase in C2-7 sagittal vertical axis were consistently observed.•Greater preoperative extension capacity corresponds to smaller decrease of C2-7 lordosis.•Range of motion reduction is larger in patients with large preoperative flexion capacity.•A large T1 slope from kyphotic thoracolumbar deformity increases lordosis reduction after laminoplasty.
Laminoplasty of the cervical spine is widely used as an effective surgical method to treat compressive myelopathy of the cervical spine; however, there is an adverse effect of kyphosis after surgery. The risk factors or predictors of kyphosis have not been sufficiently evaluated.
To assess the risk factors for kyphosis following laminoplasty.
Retrospective study.
Patients diagnosed with cervical spondylotic myelopathy (CSM) or ossification of the posterior longitudinal ligament (OPLL) who underwent laminoplasty between May 2011 and October 2018 were enrolled.
Changes in lordosis and range of motion (ROM).
Radiological imaging data were collected from simple neutral and flexion-extension radiographs at baseline and at 2-year follow-up. The ROM from the neutral position to complete flexion was defined as the flexion capacity, and the ROM from the neutral position to complete extension was defined as the extension capacity.
This study included 53 patients (mean age, 59.3 years). Multivariate linear regression analysis revealed that, the smaller the preoperative extension capacity, the greater was the decrease in lordosis (p=.025), while the larger the T1 slope, the greater was the decrease in lordosis following laminoplasty (p= .008). Correlation analysis revealed that C2-7 lordosis increased with increasing baseline T1 slope before surgery (p< .01). In patients with large preoperative C2-7 lordosis, the postoperative decrease in ROM tended to be greater (p= .028). However, the degree of lordosis and ROM reduction did not demonstrate a clear correlation with the clinical outcomes at 2 years after surgery.
Kyphotic changes in the cervical spine following laminoplasty were related to preoperative radiological parameters. The greater the preoperative extension capacity, the lower was the decrease in lordosis, and the greater the T1 slope, the greater was the decrease in lordosis.
Purpose
To evaluate feasibility of computer tomography texture analysis (CTTA) at different energy level using dual-energy spectral detector CT for liver fibrosis.
Materials and methods
Eighty-seven ...patients who underwent a spectral CT examination and had a reference standard of liver fibrosis (histopathologic findings,
n
= 61, or clinical findings for normal,
n
= 26) were included. Mean gray-level intensity, mean number of positive pixels (MPP), entropy, skewness, and kurtosis using commercially available software (TexRAD) were compared at different energy levels. Optimal CTTA parameter cutoffs to diagnose liver fibrosis were evaluated. CTTA parameters at different energy levels correlated with liver fibrosis. The association of CTTA parameters with energy level was evaluated.
Results
Mean gray-level intensity, skewness, kurtosis, and entropy showed significant differences between patients with and without clinically significant hepatic fibrosis (
P
< 0.05). Mean gray-level intensity at 50 keV was significantly positively correlated with liver fibrosis (
ρ
= 0.502,
P
< 0.001). To diagnose stages F2–F4, entropy and mean gray-level intensity at low keV level showed the largest area under the curve (AUC; 0.79 and 0.79). Estimated marginal means (EMMs) of mean gray-level intensity showed prominent differences at low energy levels.
Conclusion
CTTA parameters from different keV levels demonstrated meaningful accuracy for diagnosis of liver fibrosis or clinically significant hepatic fibrosis.
To investigate the value of ultrasound (US) microflow assessment in distinguishing malignant from benign solid breast masses as well as the association between US parameters and histologic ...microvessel density (MVD).
Ninety-eight breast masses (57 benign and 41 malignant) were examined using Superb Microvascular Imaging (SMI) and contrast-enhanced US (CEUS) before biopsy. Two radiologists evaluated the quantitative and qualitative vascular parameters on SMI (vascular index, morphology, distribution, and penetration) and CEUS (time-intensity curve analysis and enhancement characteristics). US parameters were compared between benign and malignant masses and the diagnostic performance was compared between SMI and CEUS. Subgroup analysis was performed according to lesion size. The effect of vascular parameters on downgrading Breast Imaging Reporting and Data System (BI-RADS) category 4A masses was evaluated. The association between histologic MVD and US parameters was analyzed.
Malignant masses were associated with a higher vascular index (15.1 ± 7.3 vs. 5.9 ± 5.6), complex vessel morphology (82.9% vs. 42.1%), central vascularity (95.1% vs. 59.6%), penetrating vessels (80.5% vs. 31.6%) on SMI (all,
< 0.001), as well as higher peak intensity (37.1 ± 25.7 vs. 17.0 ± 15.8,
< 0.001), slope (10.6 ± 11.2 vs. 3.9 ± 4.2,
= 0.001), area (1035.7 ± 726.9 vs. 458.2 ± 410.2,
< 0.001), hyperenhancement (95.1% vs. 70.2%,
= 0.005), centripetal enhancement (70.7% vs. 45.6%,
= 0.023), penetrating vessels (65.9% vs. 22.8%,
< 0.001), and perfusion defects (31.7% vs. 3.5%,
< 0.001) on CEUS (
≤ 0.023). The areas under the receiver operating characteristic curve (AUCs) of SMI and CEUS were 0.853 and 0.841, respectively (
= 0.803). In 19 masses measuring < 10 mm, central vascularity on SMI was associated with malignancy (100% vs. 38.5%,
= 0.018). Considering all benign SMI parameters on the BI-RADS assessment, unnecessary biopsies could be avoided in 12 category 4A masses with improved AUCs (0.500 vs. 0.605,
< 0.001). US vascular parameters associated with malignancy showed higher MVD (
≤ 0.016). MVD was higher in malignant masses than in benign masses, and malignant masses negative for estrogen receptor or positive for Ki67 had higher MVD (
< 0.05).
US microflow assessment using SMI and CEUS is valuable in distinguishing malignant from benign solid breast masses, and US vascular parameters are associated with histologic MVD.
Objectives
To explore the importance of quantitative characteristics of dual-energy CT (DECT) between pulmonary metastasis and benign lung nodules in thyroid cancer.
Methods
In this retrospective ...study, we identified 63 patients from our institution’s database with pathologically proven thyroid cancer who underwent DECT to assess pulmonary metastasis. Among these patients, 22 had 55 pulmonary metastases, and 41 had 97 benign nodules. If nodules showed increased iodine uptake on I-131 single-photon emission computed tomography-computed tomography or increased size in follow-up CT, they were considered metastatic. We compared the clinical findings and DECT parameters of both groups and performed a receiver operating characteristic analysis to evaluate the optimal cutoff values of the DECT parameters.
Results
Patients with metastases were significantly older than patients with benign nodules (
p
= 0.048). The DECT parameters of the metastatic nodules were significantly higher than those of the benign nodules (iodine concentration IC, 5.61 ± 2.02 mg/mL vs. 1.61 ± 0.98 mg/mL; normalized IC NIC, 0.60 ± 0.20 vs. 0.16 ± 0.11; NIC using pulmonary artery NIC
PA
, 0.60 ± 0.44 vs. 0.15 ± 0.11; slope of the spectral attenuation curves λHU, 5.18 ± 2.54 vs. 2.12 ± 1.39; and
Z
-effective value
Z
eff
, 10.0 ± 0.94 vs. 8.79 ± 0.75; all
p
< 0.001). In the subgroup analysis according to nodule size, all DECT parameters of the metastatic nodules in all subgroups were significantly higher than those of the benign nodules (all
p
< 0.05). The cutoff values for IC, NIC, λHU, NIC
PA
, and
Z
eff
for diagnosing metastases were 3.10, 0.29, 3.57, 0.28, and 9.34, respectively (all
p
< 0.001).
Conclusions
DECT parameters can help to differentiate metastatic and benign lung nodules in thyroid cancer.
Key Points
• DECT parameters can help to differentiate metastatic and benign lung nodules in patients with thyroid cancer.
• DECT parameters showed a significant difference between benign lung nodules and lung metastases, even for nodules with diameters ≥ 3 mm and < 5 mm.
• Among the DECT parameters, the highest diagnostic accuracy for differentiating pulmonary metastases from benign lung nodules was achieved with the NIC and IC, followed by the NIC
PA
and λHU, and their cutoff values were 0.29, 3.10, 0.28, and 3.57, respectively.
We aimed to evaluate the relationship between the capillary abnormalities including nonperfusion area (NPA) in optical coherence tomography angiography (OCTA) images and the recurrence of macular ...edema (ME) secondary to branch retinal vein occlusion (BRVO) after intravitreal injection of anti-vascular endothelial growth factor (anti-VEGF; bevacizumab). The records of 40 patients who underwent intravitreal bevacizumab injection for ME secondary to BRVO and had at least six months of follow-up were reviewed. Central retinal thickness (CRT; μm) and macular edema type were evaluated prior to treatment. After ME resolution, nonperfusion areas in the 1 mm (NPA1) and 1-3 mm (NPA3) zones on the Early Treatment Diabetic Retinopathy Study (ETDRS) circle within the superficial capillary plexus (SCP) and deep capillary plexus (DCP) were measured using OCTA images. Furthermore, other microvascular abnormalities in the both SCP and DCP were compared between groups. ME recurred in 25 of 40 (62.5%) eyes. The NPA1 of the SCP and DCP (p = 0.002, 0.004, respectively), NPA3 of the SCP and DCP (p = 0.002, 0.008, respectively), and initial CRT (p = 0.022) differed significantly between eyes with and without ME recurrence. In multivariate logistic regression analyses, the NPA1 of the DCP (OR: 344.718; p = 0.029) and NPA3 of the SCP (OR: 4.072; p = 0.018) were significantly associated with ME recurrence. Other microvascular abnormalities were not significantly different between two groups. The central NPA and parafoveal NPA of the SCP in OCTA images correlated strongly with ME recurrence in BRVO patients after intravitreal anti-VEGF injection.
No published studies have evaluated the accuracy of volumetric measurement of solid nodules and ground-glass nodules on low-dose or ultra–low-dose chest computed tomography, reconstructed using deep ...learning–based algorithms. This is an important issue in lung cancer screening. Our study aimed to investigate the accuracy of semiautomatic volume measurement of solid nodules and ground-glass nodules, using two deep learning–based image reconstruction algorithms (Truefidelity and ClariCT.AI), compared with iterative reconstruction (ASiR-V) in low-dose and ultra–low-dose settings. We performed computed tomography scans of solid nodules and ground-glass nodules of different diameters placed in a phantom at four radiation doses (120 kVp/220 mA, 120 kVp/90 mA, 120 kVp/40 mA, and 80 kVp/40 mA). Each scan was reconstructed using Truefidelity, ClariCT.AI, and ASiR-V. The solid nodule and ground-glass nodule volumes were measured semiautomatically. The gold-standard volumes could be calculated using the diameter since all nodule phantoms are perfectly spherical. Subsequently, absolute percentage measurement errors of the measured volumes were calculated. Image noise was also calculated. Across all nodules at all dose settings, the absolute percentage measurement errors of Truefidelity and ClariCT.AI were less than 11%; they were significantly lower with Truefidelity or ClariCT.AI than with ASiR-V (all P<0.05). The absolute percentage measurement errors for the smallest solid nodule (3 mm) reconstructed by Truefidelity or ClariCT.AI at all dose settings were significantly lower than those of this nodule reconstructed by ASiR-V (all P<0.05). Furthermore, the lowest absolute percentage measurement errors for ground-glass nodules were observed with Truefidelity or ClariCT.AI at all dose settings. The absolute percentage measurement errors for ground-glass nodules reconstructed with Truefidelity at ultra–low-dose settings were significantly lower than those of all sizes of ground-glass nodules reconstructed with ASiR-V (all P<0.05). Image noise was lowest with Truefidelity (all P<0.05). In conclusion, the deep learning–based algorithms were more accurate for volume measurements of both solid nodules and ground-glass nodules than ASiR-V at both low-dose and ultra–low-dose settings.
Celotno besedilo
Dostopno za:
DOBA, IZUM, KILJ, NUK, PILJ, PNG, SAZU, SIK, UILJ, UKNU, UL, UM, UPUK
The aim of this study was to develop a predictive model of objective oropharyngeal obstructive sleep apnea (OSA) surgery outcomes including success rate and apnea-hypopnea index (AHI) reduction ratio ...in adult OSA patients.
Retrospective outcome research.
All subjects with OSA who underwent oropharyngeal and/or nasal surgery and were followed for at least 3 months were enrolled in this study. Demographic, anatomical tonsil size (TS) and palate-tongue position (PTP) grade (Gr), and polysomnographic parameters were analyzed. The AHI reduction ratio (%) was defined as (postoperative AHI-preoperative AHI) x 100 / postoperative AHI, and surgical success was defined as a ≥ 50% reduction in preoperative AHI with a postoperative AHI < 20.
A total of 156 consecutive OSAS adult patients (mean age ± SD = 38.9 ± 9.6, M / F = 149 / 7) were included in this study. The best predictive equation by Forward Selection likelihood ratio (LR) logistic regression analysis was: Formula: see textThe best predictive equation according to stepwise multiple linear regression analysis was: Formula: see text (TS/PTP Gr = 1 if TS/PTP Gr 3 or 4, TS/PTP Gr = 0 if TS/PTP Gr 1 or 2).
The predictive models for oropharyngeal surgery described in this study may be useful for planning surgical treatments and improving objective outcomes in adult OSA patients.
Celotno besedilo
Dostopno za:
DOBA, IZUM, KILJ, NUK, PILJ, PNG, SAZU, SIK, UILJ, UKNU, UL, UM, UPUK