Current Medical Imaging - Volume 19, Issue 9, 2023

Amide proton transfer (APT) imaging is a technique that reflects the level of amide proton transport of hydrogen ions in water to peptides and proteins within the tumour by using magnetic resonance imaging without contrast administration. This technique can more accurately reflect the protein level, pH value, and other relevant information within the tumour. Currently, APT is widely used in glioma research. In this paper, we briefly describe the research progress on this technique.

Background: Cone Beam Computed Tomography (CBCT) represents the optimal imaging solution for the evaluation of the maxillofacial and dental area when quantitative geometric and volumetric accuracy is necessary (e.g., in implantology and orthodontics). Moreover, in recent years, this technique has given excellent results for the imaging of lower and upper extremities. Therefore, significant interest has been increased in using CBCT to investigate larger and non-traditional anatomical districts. Objective: The purpose of this work is to review the scientific literature in Pubmed and Scopus on CBCT application beyond head districts by paying attention to image quality and radiological doses. Methods: The search for keywords was conducted in Pubmed and Scopus databases with no back-date restriction. Papers on applications of CBCT to head were excluded from the present work. From each considered paper, parameters related to image quality and radiological dose were extracted. An overall qualitative evaluation of the results extracted from each issue was done by comparing the conclusive remarks of each author regarding doses and image quality. PRISMA statements were followed during this process. Results: The review retrieved 97 issues from 83 extracted papers; 46 issues presented a comparison between CBCT and Multi-Detector Computed Tomography (MDCT), and 51 reviewed only CBCT. The radiological doses given to the patient with CBCT were considered acceptable in 91% of cases, and the final image quality was found in 99%. Conclusion: CBCT represents a promising technology not only for imaging of the head and upper and lower extremities but for all the orthopedic districts. Moreover, the application of CBCT derived from C-arms (without the possibility of a 360 ° rotation range) during invasive investigations demonstrates the feasibility of this technique for non-standard anatomical areas, from soft tissues to vascular beds, despite the limits due to the incomplete rotation of the tube.

Objective: The objective of the paper was to compare the value of CT and MRI in the diagnosis of primary carcinoma of the liver. Methods: A retrospective analysis was performed on 132 cases of suspected primary liver carcinoma. CT and MRI diagnosis were performed and pathological results were compared to determine the diagnostic value of the two methods. Results: 96 cases were diagnosed as primary liver carcinoma by pathological examination after operation. The total detection rate of 96 lesions through MRI was 93.75%, while 84.38% through CT (P<0.05). For lesions with a <3 cm diameter, the CT detection rates of lesions in the plain, arterial, portal, and equilibrium phases were 52.94%, 73.53%, 58.82%, and 58.82% respectively. For lesions with a diameter ≥ 3 cm, the CT detection rate was 80.65 %, 93.55%, 85.48%, and 83.87%, respectively (P<0.05). For lesions with <3cm diameter, the MRI detection rates of lesions in the T1WI, T2WI, LAVA arterial phase, LAVA portal phase, and LAVA balance phase were 61.76%, 76.47%, 88.24%, 79.41%, and 52.94%, respectively, and for lesions with ≥3cm diameter, the detection rates of MRI were 77.42%, 87.10%, 91.94%, 90.32%, and 90.32%, respectively, and the detection rate of lesions with ≥3cm diameter in the balance phase of LAVA was higher (P<0.05). Taking pathological results as the gold standard, the sensitivity of diagnosing primary liver carcinoma through CT is 81.25%, specificity is 75.00%, accuracy is 79.55%, the positive predictive value is 89.66%, the negative predictive value is 60.00%, and the values of the same parameters for the MRI are 93.75. %, 86.11%, 91.67%, 94.74%, and 83.78% respectively. Conclusion: Both CT and MRI have diagnostic value for primary liver carcinoma. The comparison showed that MRI has a higher diagnostic value and higher detection rate for small lesions. However, the actual process of diagnosis cannot rely solely on MRI, and a comprehensive combination of diagnosis methods will be effective.

Purpose: The aim of the study was to investigate the feasibility of discriminating between clear-cell renal cell carcinoma (ccRCC) and non-clear-cell renal cell carcinoma (non-ccRCC) via radiomics models and nomogram. Methods: The retrospective study included 147 patients (ccRCC=100, non-ccRCC=47) who underwent enhanced CT before surgery. CT images of the corticomedullary phase (CMP) were collected and features from the images were extracted. The data were randomly grouped into training and validation sets according to 7:3, and then the training set was normalized to extract the normalization rule for the training set, and then the rule was applied to the validation set. First, the T-test, T'-test or Wilcoxon rank-sum test were executed in the training set data to keep the statistically different parameters, and then the optimal features were picked based on the least absolute shrinkage and selection operator (LASSO) algorithm. Five machine learning (ML) models were trained to differentiate ccRCC from noccRCC, rad+cli nomogram was constructed based on clinical factors and radscore (radiomics score), and the performance of the classifier was mainly measured by area under the curve (AUC), accuracy, sensitivity, specificity, and F1. Finally, the ROC curves and radar plots were plotted according to the five performance parameters. Results: 1130 radiomics features were extracted, there were 736 radiomics features with statistical differences were obtained, and 4 features were finally selected after the LASSO algorithm. In the validation set of this study, three of the five ML models (logistic regression, random forest and support vector machine) had excellent performance (AUC 0.9-1.0) and two models (adaptive boosting and decision tree) had good performance (AUC 0.7-0.9), all with accuracy ≥ 0.800. The rad+cli nomogram performance was found excellent in both the training set (AUC = 0.982,0.963-1.000, accuracy=0.941) and the validation set (AUC = 0.949,0.885-1.000, accuracy=0.911). The random forest model with perfect performance (AUC = 1, accuracy=1) was found superior compared to the model performance in the training set. The rad+cli nomogram model prevailed in the comparison of the model's performance in the validation set. Conclusion: The ML models and nomogram can be used to identify the relatively common pathological subtypes in clinic and provide some reference for clinicians.

Aims: The aim of the study is to demonstrate a non-invasive alternative method to aid the decision making process in the management of adrenal masses. Background: Lipid-poor adenomas constitute 30% of all adrenal adenomas. When discovered incidentally, additional dynamic adrenal examinations are required to differentiate them from an adrenal malignancy or pheochromocytoma. Objective: In this retrospective study, we aimed to discriminate lipid-poor adenomas from other lipidpoor adrenal masses by using radiomics analysis in single contrast phase CT scans. Materials and Methods: A total of 38 histologically proven lipid-poor adenomas (Group 1) and 38 cases of pheochromocytoma or malignant adrenal mass (Group 2) were included in this retrospective study. Lesions were segmented volumetrically by two independent authors, and a total of 63 sizes, shapes, and first- and second-order parameters were calculated. Among these parameters, a logit-fit model was produced by using 6 parameters selected by the LASSO (least absolute shrinkage and selection operator) regression. The model was cross-validated with LOOCV (leave-one-out crossvalidation) and 1000-bootstrap sampling. A random forest model was also generated in order to use all parameters without the risk of multicollinearity. This model was examined with the nested crossvalidation method. Results: Sensitivity, specificity, accuracy and AUC were calculated in test sets as 84.2%, 81.6%, 82.9% and 0.829 in the logit fit model and 91%, 80%, 82.8% and 0.975 in the RF model, respectively. Conclusion: Predictive models based on radiomics analysis using single-phase contrast-enhanced CT can help characterize adrenal lesions.

Background: Delphian lymph node (DLN) has been considered to be a gate that predicts widespread lymph node involvement, higher recurrence and mortality rates of head and neck cancer. Objective: This study aimed to establish a preoperative ultrasonography integrated machine learning prediction model to predict Delphian lymph node metastasis (DLNM) in patients with diagnosed papillary thyroid carcinoma (PTC). Methods: Ultrasonographic and clinicopathologic variables of PTC patients from 2014 to 2021 were retrospectively analyzed. The risk factors associated with DLNM were identified and validated through a developed random forest (RF) algorithm model based on machine learning and a logistic regression (LR) model. Results: A total of 316 patients with 402 thyroid lesions were enrolled for the training dataset and 280 patients with 341 lesions for the validation dataset, with 170 (28.52%) patients developed DLNM. The elastography score of ultrasonography, central lymph node metastasis, lateral lymph node metastasis, and serum calcitonin were predictive factors for DLNM in both models. The RF model has better predictive performance in the training dataset and validation dataset (AUC: 0.957 vs. 0.890) than that in the LR model (AUC: 0.908 vs. 0.833). Conclusion: The preoperative ultrasonography integrated RF model constructed in this study could accurately predict DLNM in PTC patients, which may provide clinicians with more personalized clinical decision-making recommendations preoperatively. Machine learning technology has the potential to improve the development of DLNM prediction models in PTC patients.

Background: Parkinsonism is a term used for the collection of clinical features that cause movement disorders similar to those in Parkinson’s disease. Accurate differentiation of these disorders is critical for the treatment and prognosis of any disease. Fluorine-18 N-(3-fluoropropyl)-2β- carboxymethoxy-3β-(4-iodophenyl) nortropane (F-18 FP-CIT) has been used in the evaluation of parkinsonism by its uptake in the dopamine active transporter (DAT) of the striatum. Its uptake in other areas of the brain, such as serotonin transporter (SERT) in the midbrain or thalamus, is also recognised. Objective: To investigate whether midbrain SERT uptake of F-18 FP-CIT on positron emission tomography (PET) could be applied to the differentiation of parkinsonism in combination with striatal DAT uptake. Methods: This retrospective study included clinically diagnosed three essential tremors (ET), 53 parkinsonism patients (21 idiopathic Parkinson’s disease (IPD), 6 multiple system atrophy – cerebellar type (MSA-C), 7 multiple system atrophy - parkinsonian type (MSA-P), 8 vascular parkinsonism (VP), and 11 drug-induced parkinsonism (DIP)), and 16 healthy controls. The patient group consisted of 29 men and 27 women (age mean ± SD years, 69.9 ± 8.5 and 69.2 ± 8.9, respectively), and the healthy controls consisted of 8 men and 8 women (age mean ± SD years, 64.5 ± 8.2 and 64.3 ± 7.6, respectively). Mean standardized uptake values (SUVs) and activity volumes were measured from the visualized FP-CIT uptake of the midbrain (substantia nigra and dorsal raphe nucleus) as well as the striatum (caudate nucleus and putamen). The mean SUVs of the occipital region were measured as the background activity. The semiquantitative binding ratio (BR) was calculated using the following formula: BR = (SUVmean of the region of interest − SUVmean of background)/SUVmean of the background. SUV, volume, and BR in each type of parkinsonism were compared with those in healthy controls using both nonparametric and parametric methods. The correlation between the visual score of the qualitative analysis and the BR was examined. Results: Except for the dorsal raphe nucleus in VP, the midbrain BRs in all parkinsonism showed a statistically significant decrease compared to those in healthy controls. Both midbrain and striatal BRs were significantly decreased only in patients with IPD or MSA-P; a greater decrease of substantia nigra BR was identified in MSA-P than in IPD (p < 0.05). The striatal BRs in MSA-C, VP, and DIP showed no significant difference from those in healthy controls. Finally, four patterns of uptake were identified: 1) decreased striatal and midbrain uptake for IPD and MSA-P, 2) normal striatal uptake and decreased midbrain uptake (both substantia nigra and dorsal raphe nucleus) for MSA-C and DIP, 3) normal striatal uptake and decreased substantia nigra uptake (without decreased dorsal raphe nucleus uptake) for VP, and 4) normal striatal and midbrain uptake for ET. Conclusion: The possible differential diagnoses were split into two groups when only striatal uptake was considered but they were divided into four groups after adding midbrain uptake. Although additional midbrain F-18 FP-CIT uptake still could not make a final definitive diagnosis, it could provide another piece of information and specific diagnostic guidelines for the differentiation of parkinsonism.

Objective: We assessed the feasibility of spectral CT imaging for diagnosing lung adenocarcinomas manifesting as ground-glass nodules (GGNs) with varying Ki-67 expression. Methods: Spectral CT parameters in 116 patients with lung adenocarcinomas manifesting as GGNs were analyzed. Cases were grouped into pre-invasive/minimally invasive adenocarcinoma (pre/MIA) and invasive adenocarcinoma (IA) groups. The Ki-67 labeling index (Ki-67 LI) was measured and compared between the two groups. Ki-67 LI was divided into three grades based on the number of positive-stained cells. Spectral CT parameters (diameter, water, and iodine concentrations of the lesion [WCL and ICL], the slope of the spectral Hounsfield unit curve [λHU], and CT values from 40 to 140 keV [at 10 keV intervals]) were compared among the three different grades. The correlation between quantitative spectral CT imaging parameters and Ki-67 LI was analyzed using Spearman correlation analysis. Results: Ki-67 LI in the IA group was significantly higher than in the pre/MIA group (p < 0.01). Grade 2 had higher diameter, WCL, and monochromatic CT values, and grade 1 had higher ICL and λHU. The WCL and monochromatic CT values were highly and positively correlated with Ki-67 LI. CT40keV had the highest correlation with Ki-67 LI, the diameter was moderately correlated with Ki- 67 LI, and ICL and λHU were weakly correlated with Ki-67 LI. Conclusion: Spectral CT, a noninvasive diagnostic method, is valuable for predicting Ki-67 expression higher in IA, thus allowing preoperative evaluation of lung adenocarcinomas manifesting as GGNs.

Objective: The present study aimed to improve the diagnosis and treatment outcome of cerebrospinal fluid (CSF) rhinorrhea caused by patent meningoencephalocele of Sternberg’s canal of the sphenoid sinus by analyzing the clinical data and imaging features of two rare cases of this disease and by reviewing the relevant literature for possible etiology, diagnoses, and treatments. Methods: Together with the relevant literature, we retrospectively studied the clinical and imaging data of two patients (mother and child) with CSF rhinorrhea caused by patent meningoencephalocele of Sternberg’s canal of the sphenoid sinus, analyzed their diagnostic and treatment procedures, and proposed a potential, feasible treatment method. Results: On the 2^nd day after surgery, the expansive sponge and iodoform gauze in the nasal cavity were removed in both patients, and the lumbar subarachnoid drainage was removed 3 days after the operation, as no nasal discharge was observed. One week after the operation, head magnetic resonance imaging (MRI) showed that the abnormal tissue in the sphenoid sinus had disappeared, and no accumulation of the CSF was observed. Both patients were discharged after 2 weeks. At the time of discharge, both patients were without nasal drip, fever, headache, and other discomforts, and they had grade 5 muscle strength in their extremities, with normal muscle tension. Conclusion: CSF rhinorrhea is usually caused by secondary factors. Spontaneous CSF rhinorrhea caused by encephalocele of the skull base due to congenital dysplasia of the skull base is very rare and easily misdiagnosed. The presence of brain tissue or CSF signal in the sphenoid sinus on preoperative MR images is an important imaging feature of the disease. Conditional cisternography can be used to further detect CSF leaks. Endoscopic transnasal transsphenoidal repair of CSF leaks combined with short-term postoperative lumbar subarachnoid drainage is an effective treatment method. According to previous literature, the possible causes of meningoencephalocele with patent Sternberg’s canal of the sphenoid sinus include abnormal development of the sphenoid sinus or the craniopharyngeal canal and bone defects of the skull base. There are no related reports on patent meningoencephalocele caused by Sternberg’s canal in direct blood relatives, such as mother-son; therefore, the possibility of this disease having a genetic origin should be considered in future studies on its pathophysiological mechanisms.

Background: Spinal radiographic analyses are valid and reliable practices used for patient management in healthcare. Technologies and tools used for these analyses need to be valid and reliable. Objective: This study investigates repeated accuracy for validity and intra- and inter-examiner reliability of computer-aided lateral spinal radiograph measurements using PostureRay^® EMR software. Methods: Rectangle renderings representing lateral spinal radiographic views were created to remove examiner interpretation of anatomical landmarks as performed on actual radiographs to examine the objective repeated accuracy of PostureRay^® software. A blind, randomized, repeated measure design was performed using four examiners trained and experienced with the Harrison Posterior Tangent method of measurement. Each examiner analyzed the spine renderings using a hand-drawn and PostureRay ^® method multiple times. A fully crossed design whereby examiners performed multiple assessments per rendering using each analysis method was used to assess intra- and inter-examiner reliability within each method and a formal comparison of the two methods. Results: Using hand-drawn and PostureRay^® methods, examiners displayed very high intra-examiner reliability with correlation values greater than 0.999 for each combination of trials. PostureRay^® method showed consistent accuracy of measurements. Further, there was no statistical difference in average concordance between hand-drawn and PostureRay^® methods for the lateral cervical, thoracic, and thoracolumbosacral spinal regions. Conclusion: This study reports a very high degree of intra- and inter-examiner reliability of radiographic line drawing methods and establishes concurrent validity of PostureRay^® EMR software in determining angles and displacements of lateral spinal alignment as an equivalent method to the handdrawn method.

Background: Haglund’s syndrome is a common cause of heel pain but often neglected clinically. Haglund’s syndrome refers to a series of symptoms caused by impingement among posterosuperior prominence of the calcaneus, bursa and Achilles tendon. It is difficult to distinguish Haglund’s syndrome from other causes of heel pain by clinical diagnosis. Imageology is of great value in the diagnosis of Haglund’s syndrome. Objective: Our study aims to summarize the Magnet resonance (MR) imaging characteristics of Haglund’s syndrome and provide some reference to clinical work. Method: We retrospectively analyzed the MR images of 11 patients (6 males; 5 females; 6 right ankles, 4 left ankles, 1 bimalleolar ankles) who have been clinically and radiologically confirmed Haglund’s syndrome. Observation contents: morphological changes of calcaneus and talus, abnormal signal of calcaneus, abnormal Achilles tendon, and soft tissue abnormalities around Achilles tendon. Combined with literature reviews, summarize the MR imaging features of Haglund’s syndrome. Results: In 12 ankles, all ankles showed posterosuperior prominence of the calcaneus and Achilles tendon degeneration; 7 ankles showed bone marrow edema; 6 Achilles tendons were graded as either type II or type III tendinosis; 5 Achilles tendons showed partial tear; 12 ankles showed retrocalcaneal bursitis, 7 ankles showed retro-Achilles bursitis, 6 ankles showed Kager’s fat pad edema. Conclusion: This study found that MR images of Haglund's syndrome showed bone edema of the calcaneus, degeneration and partial tear of the Achilles tendon, the retrocalcaneal and retro-Achilles bursas, and Kager’s fat pad edema.

Background: Diffusion tensor imaging (DTI), a novel method of describing nerve structure, is a special form of magnetic resonance imaging (MRI). This new imaging method can be used to locate the diseased nerve roots in lumbar disc herniation. Objective: The objective of this study is to compare patient outcomes between single-level and doublesection percutaneous endoscopic lumbar discectomy (PELD) in the treatment of lumbar intervertebral disc herniation with single nerve root compression, where single-sided PELD is guided by magnetic resonance diffusion tensor imaging (DTI). Methods: The clinical data of patients with lumbar intervertebral disc herniation with double compression of single nerve root symptoms in the Affiliated Hospital of Weifang Medical University from January 2019 to May 2021 were retrospectively summarized and divided into single-level percutaneous endoscopic discectomy (PELD) group after DTI localization and double-section PELD group. The operation time, intraoperative bleeding, VAS score and JOA score of the two groups were compared, as well as the preoperative and postoperative fractional anisotropy (FA) values in the DTI group. Results: The operation time and intraoperative bleeding volume of patients in the DTI group were significantly lower than those in the double segment group, and there was no significant difference between VAS scores and JOA scores in the two groups. After the operation, the nerve root FA value of the responsible compression site of patients in the DTI group increased significantly, but it was still lower than the healthy symmetrical part. Conclusion: The single-level PELD based on DTI has achieved a similar effect to that of the doublesegment PELD in 3 months after the operation, which can improve the FA value of the lesion nerve root. Its bleeding amount is less, and the operation time is shorter, but the efficacy of this technology still requires long-term follow-up of large samples.

Background: In abdominal magnetic resonance imaging (MRI), the late hepatic arterial phase is particularly important for the diagnosis of hepatocellular carcinoma (HCC). However, poor patient compliance with breath-hold imaging acquisition protocols and the administration of the liverspecific contrast agent gadolinium ethoxybenzyl-diethylenetriaminepentaacetic acid (Gd-EOB-DTPA) increases the motion artifacts that degrade the quality of the images making it more difficult to interpret the images. Objective: This study aimed to evaluate the role of adaptive respiratory training in improving the quality of the MRI by reducing the motion artifacts and improving the signal intensity from the regions of interest (ROI) in the late hepatic arterial phase. Methods: A total of 120 patients who underwent an abdominal MRI between 2021 to 2022 to assess for the liver disease were included in the study. These patients were divided into two groups: the experimental group and the control group. The patients in the experimental group received adaptive training. The incidence of motion artifacts and the signal intensity in the late hepatic arterial phase within the abdominal aorta, hepatic artery, splenic artery, and hepatic parenchyma between the experimental group and the control group were compared. Results: The incidence of motion artifacts in the experimental group was significantly reduced by 28.3% (p = 0.001, Chi-square value = 12.079). In the late hepatic arterial phase, the signal intensity of the abdominal aorta, the hepatic artery, the splenic artery, and the hepatic parenchyma increased by 7.3%, 27.4%, 29.5%, and 6.9%, respectively. Conclusion: Adaptive respiratory training reduced the incidence of motion artifacts and improved the signal intensity for various ROI. The improved image quality could potentially facilitate image interpretation and reduce the number of repeat MRI scans.

Background: In patients treated with curative surgery for colorectal cancer, the single-phase (portal phase) abdomen-pelvic computed tomography (CT) follow-up is routinely performed to detect any recurrent lesions. Objective: The objective of this study is to evaluate CT features implying metastasis in newly appearing hepatic lesions on single-phase CT after colorectal cancer surgery. Methods: Of 1,227 patients who underwent curative colorectal surgery between 2011 and 2014, 267 patients with new hepatic lesions on follow-up single-phase CT (metastasis, n=192; benign, n=75) were included. The CT features of the hepatic lesions were analyzed by two radiologists retrospectively, and clinical information, including carcinoembryonic antigen (CEA) level was also reviewed. The risk factors for hepatic metastasis were evaluated by multivariable logistic regression. Results: In this study, independent CT features associated with metastasis were spherical shape (p < 0.001), circumscribed margin (p = 0.021), heterogeneous attenuation (p = 0.024) and bulging appearance (p = 0.007). Among the clinical factors, increased CEA level (p = 0.013), higher preoperative stage (p = 0.014) and lymphovascular invasion (p = 0.003) were associated with metastasis. However, on multivariable analysis, spherical shape was the only significant predictive factor for hepatic metastasis. The diagnostic performance of ‘spherical shape’ for detection of hepatic metastasis showed an accuracy of 84.6%, sensitivity of 83.9%, and specificity of 86.7%. Conclusion: Specific single-phase CT features could be useful in distinguishing hepatic metastasis from benign, resulting in continuing routine CT follow-up without liver magnetic resonance imaging (MRI) when metastasis is unlikely.