Current Medical Imaging - Volume 16, Issue 8, 2020

Background: This paper endeavors to identify an expedient approach for the detection of the brain tumor in MRI images. The detection of tumor is based on i) review of the machine learning approach for the identification of brain tumor and ii) review of a suitable approach for brain tumor detection. Discussion: This review focuses on different imaging techniques such as X-rays, PET, CT- Scan, and MRI. This survey identifies a different approach with better accuracy for tumor detection. This further includes the image processing method. In most applications, machine learning shows better performance than manual segmentation of the brain tumors from MRI images as it is a difficult and time-consuming task. For fast and better computational results, radiology used a different approach with MRI, CT-scan, X-ray, and PET. Furthermore, summarizing the literature, this paper also provides a critical evaluation of the surveyed literature which reveals new facets of research. Conclusion: The problem faced by the researchers during brain tumor detection techniques and machine learning applications for clinical settings have also been discussed.

Deep learning has attracted great attention in the medical imaging community as a promising solution for automated, fast and accurate medical image analysis, which is mandatory for quality healthcare. Convolutional neural networks and its variants have become the most preferred and widely used deep learning models in medical image analysis. In this paper, concise overviews of the modern deep learning models applied in medical image analysis are provided and the key tasks performed by deep learning models, i.e. classification, segmentation, retrieval, detection, and registration are reviewed in detail. Some recent researches have shown that deep learning models can outperform medical experts in certain tasks. With the significant breakthroughs made by deep learning methods, it is expected that patients will soon be able to safely and conveniently interact with AI-based medical systems and such intelligent systems will actually improve patient healthcare. There are various complexities and challenges involved in deep learning-based medical image analysis, such as limited datasets. But researchers are actively working in this area to mitigate these challenges and further improve health care with AI.

Cardiac amyloidosis (CA) is an infiltrative disease characterized by the extracellular deposition of fibrils, amyloid, in the heart. The vast majority of patients with CA has one of two types between transthyretin amyloid (ATTR) and immunoglobulin light chain associated amyloid (AL), that have different prognosis and therapeutic options. CA is often underdiagnosed. The histological analysis of endomyocardial tissue is the gold standard for the diagnosis, although it has its limitations due to its invasive nature. Nuclear medicine now plays a key role in the early and accurate diagnosis of this disease, and in the ability to distinguish between the two forms. Recent several studies support the potential advantage of bone-seeking radionuclides as a screening technique for the most common types of amyloidosis, in particular ATTR form. This review presents noninvasive modalities to diagnose CA and focuses on the radionuclide imaging techniques (bone-seeking agents scintigraphy, cardiac sympathetic innervation and positron emission tomography studies) available to visualize myocardial amyloid involvement. Furthermore, we report the case of an 83-year old male with a history of prostate cancer, carcinoma of the cecum and kidney cancer, submitted to bone scan to detect bone metastasis, that revealed a myocardial uptake of 99mTC-HMPD suggestive of ATTR CA. An accurate and early diagnosis of CA able to distinguish beyween AL and ATTR CA combined to the improving therapies could improve the survival of patients with this disease.

Three-dimensional (3D) printing is increasingly used in medical applications with most of the studies focusing on its applications in medical education and training, pre-surgical planning and simulation, and doctor-patient communication. An emerging area of utilising 3D printed models lies in the development of cardiac computed tomography (CT) protocols for visualisation and detection of cardiovascular disease. Specifically, 3D printed heart and cardiovascular models have shown potential value in the evaluation of coronary plaques and coronary stents, aortic diseases and detection of pulmonary embolism. This review article provides an overview of the clinical value of 3D printed models in these areas with regard to the development of optimal CT scanning protocols for both diagnostic evaluation of cardiovascular disease and reduction of radiation dose. The expected outcomes are to encourage further research towards this direction.

Background: Epigenetic dysfunction is implicated in many neurologic, psychiatric and oncologic diseases. Consequently, histone deacetylases (HDACs) inhibitors have been developed as therapeutic and imaging agents for these diseases. However, only a few radiotracers have been developed as HDACs imaging agents for the central nervous system (CNS). We report herein the synthesis and evaluation of [¹⁸F]INER-1577-3 ([¹⁸F]5) as an HDACs imaging agent for CNS. Methods: [¹⁸F]INER-1577-3 ([¹⁸F]5) was synthesized by two methods: one-step (A) and two-step (B) methods. Briefly, radiofluorination of the corresponding precursors (11, 12) with K[¹⁸F]/K2.2.2 followed by purifications with HPLC gave ([¹⁸F]5). The quality of [¹⁸F]INER- 1577-3 synthesized by these methods was verified by HPLC and TLC as compared to an authentic sample. The inhibitions of [¹⁸F]INER-1577-3 and related HDACs inhibitors on tumor cells growth were carried out with breast cancer cell line 4T1 and MCF-7. The whole-body and brain uptake of [¹⁸F]INER-1577-3 in rats and AD mice were determined using a micro-PET scanner and the data was analyzed using PMOD. Results: The radiochemical yield of [¹⁸F]INER-1577-3 synthesized by these two methods was 1.4 % (Method A) and 8.8% (Method B) (EOB), respectively. The synthesis time was 115 min and 100 min, respectively, from EOB. The inhibition studies showed that INER-1577-3 has a significant inhibitory effect in HDAC6 and HDAC8 but not HDAC2. PET studies in rats and AD mice showed a maximum at about 15 min postinjection for the whole brain of a rat (0.47 ± 0.03 %ID/g), SAMP8 mice (5.63 ± 1.09 %ID/g) and SAMR1 mice (7.23 ± 1.21 %ID/g). Conclusion: This study showed that INER-1577-3 can inhibit tumor cell growth and is one of a few HDACs inhibitors that can penetrate the blood-brain barrier (BBB) and monitor HDAC activities in AD mice. Thus, [¹⁸F]INER-1577-3 may be a potent HDACs imaging agent, especially for CNS.

Background: Quantitative susceptibility mapping (QSM) technique had been used to measure the magnetic susceptibility of brain tissue in clinical practice. However, QSM presented echo-time (TE) dependence, and an appropriate number of echo-times (nTEs) for QSM became more important to obtain the reliable susceptibility value. Objective: The aim of the study was to explore the optimal nTEs for quantitative susceptibility mapping (QSM) measurements of basal ganglia nuclei in the healthy brain. Methods: 3D multi-echo enhanced gradient recalled echo T2 star weighted angiography (ESWAN) sequence was acquired on a 3.0T MR scanner for QSM analysis. Regions of interests (ROIs) were drawn along the margin of the head of the caudate nucleus (HCN), putamen (Pu) and globus pallidus (GP). The mean susceptibility value and standard deviation of the ROIs were derived from the pixels within each region. Results: CV analysis demonstrated that TE6, TE8 and TE14 ESWAN sequences presented consistent lower CV value (< 1) for QSM measure of HCN, Pu and GP. ANOVA identified that susceptibility value showed no significant difference between TE6 and TE8 in HCN, Pu and GP (P > 0.05). ICC analysis demonstrated that the susceptibility value of TE6-TE8 had the highest ICC value as compared with TE6-TE14 and TE8-TE14 in HCN, Pu and GP. Combined with the timeefficiency of MRI scanning, TE6 sequence could not only provide the reliable QSM measurement but also short imaging time. Conclusion: The current study identified that the optimal nTEs of ESWAN were 6 TEs (2.9ms ~ 80.9ms) for QSM measurement of basal ganglia nuclei in the healthy brain.

Background: Accurate localization of the lumpectomy cavity is important for breast cancer radiotherapy after breast-conserving surgery (BCS), but the LC localization based on CT is often difficult to delineate accurately. The study aimed to compare CT-defined LC planning to MRI-defined findings in the supine position for higher soft-tissue resolution of MRI. Methods: Fifty-nine breast cancer patients underwent radiotherapy CT planning in supine position followed by MR imaging on the same day. LC was contoured by the radiologist and radiation oncologist together by CT and MRI separately. T2 weighted MR images and tomography findings were combined and the LC volume, mean diameter and the longest axis length were measured after contouring. Subsequently, patients were divided into two groups according to seroma in LC and the above-mentioned parameters were compared. Results: We did not find any statistically significant difference in the LC volume, mean diameter and length at the longest axis between CT and MRI but based on the presence or absence of seroma, statistically significant differences were found in the LC volumes and the length at the longest axis of LC volumes. Conclusion: We believe that the supine MRI in the same position with CT will be more effective for radiotherapy planning, particularly in patients without a seroma in the surgical cavity.

Objective: False-positive nodule reduction is a crucial part of a computer-aided detection (CADe) system, which assists radiologists in accurate lung nodule detection. In this research, a novel scheme using multi-level 3D DenseNet framework is proposed to implement false-positive nodule reduction task. Methods: Multi-level 3D DenseNet models were extended to differentiate lung nodules from falsepositive nodules. First, different models were fed with 3D cubes with different sizes for encoding multi-level contextual information to meet the challenges of the large variations of lung nodules. In addition, image rotation and flipping were utilized to upsample positive samples which consisted of a positive sample set. Furthermore, the 3D DenseNets were designed to keep low-level information of nodules, as densely connected structures in DenseNet can reuse features of lung nodules and then boost feature propagation. Finally, the optimal weighted linear combination of all model scores obtained the best classification result in this research. Results: The proposed method was evaluated with LUNA16 dataset which contained 888 thin-slice CT scans. The performance was validated via 10-fold cross-validation. Both the Free-response Receiver Operating Characteristic (FROC) curve and the Competition Performance Metric (CPM) score show that the proposed scheme can achieve a satisfactory detection performance in the falsepositive reduction track of the LUNA16 challenge. Conclusion: The result shows that the proposed scheme can be significant for false-positive nodule reduction task.

Background: With the evaluation of focal epicardial coronary stenosis and non-obstructive atherosclerosis, the cardiac scans play a significant role in diagnosing coronary artery disease (CAD). Moreover, the advancements in the imaging techniques leading to improved risk assessment and timely therapies help in early diagnosis of CAD with greater accuracy. Aims: To evaluate the role of cardiac scan in diagnosing CAD. Methods: Recruited 100 individuals without any history of CAD that refers to the assessment of suspected angina, conducted the prospective study. Electrocardiogram (ECG) findings assisted in the evaluation of left bundle branch blockage, abnormalities of ST-segment, and pathological Q waves. Results: The results depicted negative N.M findings among 38 respondents; whereas, ischemia and myocardial infarctions were diagnosed in 26% and 19% of the respondents, respectively. The majority of the males (59) were positive in contrast to 37 females with positive results. Similarly, 24 respondents were presented with mild dilated left atria (LA), 37 respondents suffered from impaired relaxation pattern of left ventricular (LV) diastolic filling; while, 40 of the respondents had normal global LV systolic function. Conclusion: The study results have concluded that non-invasive, low-risk, and cost-effective technique like ECG is an important beneficial advancement in the diagnosis of CAD.

Background: Initial staging of rectal cancer is done by high-resolution magnetic resonance imaging (MRI), however, pelvic computed tomography (CT) is also frequently used. The aim of this study was to evaluate the added clinical benefit of pelvic CT or whether it can alter the initial staging or not. Methods: The study was composed of 187 patients with rectal cancer. Firstly, imaging except pelvic CT was evaluated. Secondly, the pelvic CT was evaluated and the staging was adjusted according to the new findings. Subsequently, the two staging results were compared to investigate the added clinical benefit of pelvic CT. Results: Compared with the imaging data except pelvic CT, new findings revealed by the pelvic CT included metastases of the pelvic bone (n = 1) and pelvic peritoneum (n = 3). However, the new findings did not change the primary staging. Of the three patients with pelvic peritoneal metastasis, two were already determined with peritoneal involvement and ascites by abdominal CT, and the third patient was observed with liver and distant lymph node metastasis. Thus, none of their initial stagings needed to be changed. Conclusion: The addition of pelvic CT to the pre-treatment imaging strategy cannot provide added clinical benefit for the primary evaluation of rectal cancer.

Introduction: With the advancement in internet technology, a large amount of information in the form of data and image is transferred from one end to the other. The information may be military, defense, medical, etc. which should be kept confidential by providing security. Objective: The aim of this article will be to provide security to the image. This is achieved by applying the image encryption method which converts the original information into an unreadable format. Methods: This work explores an efficient way of image encryption using a chaotic logistic function. A set of two chaotic logistic functions and 256 bit long external secret key are employed to enhance the security in the encrypted images. The initial condition of first logistic function has been obtained by providing the suitable weights to all bits of the secret key. The initial condition of second logistic function has been derived from the first chaotic logistic function. In this proposed algorithm, ten different operations are used to encrypt the pixel of an image. The outcome of the second logistic map decides the operation to be used in the encryption of the particular image pixel. Results: Various statistical parameters like NPCR, UACI and information entropy were calculated. Conclusion: Results show that the proposed algorithm provides an image encryption method with better security and efficiency for all real-time applications such as medical images.

Backround: Less than 1% of all intracranial tumors are in the pineal region. Papillary tumor of the pineal region is considered one of four pineal parenchymal tumors according to the 2007 World Health Organisation classification of central nervous system tumors. It is mostly seen in the middle age and it is rare under the age of 17 in the literature. In this report, we aim to present a case with papillary tumor of the pineal region with both CT and MRI findings, and discuss the differential diagnosis. Case report: A 17-year-old female patient who underwent a treatment of ventriculoperitoneal shunt due to hydrocephalus 18 months ago applied to the neurosurgery clinic with nausea, vomiting, sensorineural hearing loss and Parinaud syndrome continuing through 1 month. Computed tomography (CT) and magnetic resonance imaging (MRI) have been applied. A 58x31x38 mm mass with cystic and mild hyperdense soft tissue components was observed in the localization of pineal gland and posterior comissure on non-enhanced CT. MRI examination revealed that the mass filled the pineal region, contained solid and large cystic components, and hyperintense on T1-weighted images. After intravenous contrast media injection, the mass enhanced more peripherally. DW-MRI showed that there is a diffusion restriction in some parts of the soft tissue components. The mass was totally resected, and pathology report revealed that it is a papillary tumor of the pineal region. Conclusion: When a T1-weighted hyperintense and CT hyperdense mass with cystic and solid components is seen in the pineal region, even if it is rare, papillary tumor should also be considered in the differential diagnosis after exclusion of other hyperintense lesions in T1-weighted imaging.

Background: Adenofibroma is a benign tumor composed histologically of epithelial elements and mesenchymal stroma. Carcinosarcoma is a malignant neoplasm that contains elements of carcinoma and sarcoma. Carcinosarcoma arising from adenofibroma of the uterus has never been reported in the literature in English language. Case Presentaion: We report a case of a 56-year-old woman who complained vaginal spotting persisting for several months. We described here for the first time a case of malignant transformation of uterine endometrial adenofibroma into carcinosarcoma that was depicted as a multilocular cystic lesion with enhancing solid portions and stalk-like structure between the myometrium and endometrial mass.

Background: Middle ear ventilation and the drainage of secretions are the basic functions of the eustachian tube. Eustachian tube (ET) dysfunction can lead to otitis media. Multiple causes of ET dysfunction exist; ET obstruction is one of them. Osseous obliteration of ET is quite rare. In the literature, there are two cases of solitary osteoma and a case of hamartoma with an osseous content in the ET. Herein, this report describes the case of a patient with multiple osseous lesions consistent with osteomas in the ET, causing ipsilateral otitis media with CT findings. Case Report: A 53-year-old woman presented with left-sided intermittent otalgia and otorrhea. Temporal bone computed tomography showed opacification of middle ear consistent with otitis media and nearly complete blockage of the left ET lumen with multiple osseous lesions wall probably consistent with osteomas on the left side. Conclusion: There are limited data about the radiological features of ET osteoma due to its rare occurrence. Radiographically, external ear osteoma generally appears as a single pedunculated hyperdense mass on CT scan originating from the tympanosquamous or tympanomastoid suture line and extending into the auditory canal with otherwise preserved canal dimensions. Considering this information, osseous lesions arising from ET bone wall and extending into the lumen in the present case were considered as osteomas. Especially in the case of unilateral chronic otitis media, ET should not remain a neglected anatomical area.

Background: The ectopic posterior pituitary (EPP) is a rare condition characterized by the ectopic location of the posterior pituitary lobe associated with varying degrees of stalk anomalies. The arachnoid cysts (AC) are benign lesions of the arachnoid, which account for 1% of all intracranial space-occupying lesions. Sellar/suprasellar ACs account for approximately 1% of all ACs. This is the first case of coexistence EPP with sellar/suprasellar AC. Case Report: A 67-year-old woman presented with 6 months history of fatigue. Her medical history was positive for irregular menstruation. Her endocrine examinations indicated low free thyroxine level with low TSH level, low oestradiol with low gonadotrophin level, slightly elevated prolactin level. Her Insulin-like growth factor-1 was below the normal levels. Dynamic contrast hypophysis MRI revealed a sellar cystic lesion with a dimension of 18 x 14 x 14 mm, extending from the suprasellar cistern, traversing the diaphragma sellae and reaching the level of the floor of the 3rd ventricle, consistent with sellar/suprasellar AC. There was no wall enhancement. The optic chiasm was compressed. The precontrast T1-weighted magnetic resonance images did not demonstrate the characteristic bright spot of posterior pituitary within the sella, which was higher in position, in the region of the median eminence. The pituitary stalk was not present. Conclusion: Although speculative, we have a hypothesis to explain how the EPP and sellar/- suprasellar AC coexist in this patient. Due to the absence of stalk, CSF may enter the sella tursica from the central aperture of the diaphragma sellae through which normally the stalk passes.