Selection of Convolutional Neural Network Model for Bladder Tumor Classification of Cystoscopy Images and Comparison with Humans.

Purpose: An investigation of various convolutional neural network (CNN)-based deep learning algorithms was conducted to select the appropriate artificial intelligence (AI) model for calculating the diagnostic performance of bladder tumor classification on cystoscopy images, with the performance of the selected model to be compared against that of medical students and urologists. Methods: A total of 3,731 cystoscopic images that contained 2,191 tumor images were obtained from 543 bladder tumor cases and 219 normal cases were evaluated. A total of 17 CNN models were trained for tumor classification with various hyperparameters. The diagnostic performance of the selected AI model was compared with the results obtained from urologists and medical students by using the receiver operating characteristic (ROC) curve graph and metrics. Results: EfficientNetB0 was selected as the appropriate AI model. In the test results, EfficientNetB0 achieved a balanced accuracy of 81%, sensitivity of 88%, specificity of 74%, and an area under the curve (AUC) of 92%. In contrast, human-derived diagnostic statistics for the test data showed an average balanced accuracy of 75%, sensitivity of 94%, and specificity of 55%. Specifically, urologists had an average balanced accuracy of 91%, sensitivity of 95%, and specificity of 88%, while medical students had an average balanced accuracy of 69%, sensitivity of 94%, and specificity of 44%. Conclusions: Among the various AI models, we suggest that EfficientNetB0 is an appropriate AI classification model for determining the presence of bladder tumors in cystoscopic images. EfficientNetB0 showed the highest performance among several models and showed high accuracy and specificity compared to medical students. This AI technology will be helpful for less experienced urologists or nonurologists in making diagnoses. Image-based deep learning classifies bladder cancer using cystoscopy images and shows promise for generalized applications in biomedical image analysis and clinical decision making.

L-serine restored lysosomal failure in cells derived from patients with BPAN reducing iron accumulation with eliminating lipofuscin.

Defective mitochondria and autophagy, as well as accumulation of lipid and iron in WDR45 mutant fibroblasts, is related to beta-propeller protein-associated neurodegeneration (BPAN). In this study, we found that enlarged lysosomes in cells derived from patients with BPAN had low enzyme activity, and most of the enlarged lysosomes had an accumulation of iron and oxidized lipid. Cryo-electron tomography revealed elongated lipid accumulation, and spectrometry-based elemental analysis showed that lysosomal iron and oxygen accumulation superimposed with lipid aggregates. Lysosomal lipid aggregates superimposed with autofluorescence as free radical generator, lipofuscin. To eliminate free radical stress by iron accumulation in cells derived from patients with BPAN, we investigated the effects of the iron chelator, 2,2'-bipyridine (bipyridyl, BIP). To study whether the defects in patient-derived cells can be rescued by an iron chelator BIP, we tested whether the level of iron and reactive oxygen species (ROS) in the cells and genes related to oxidative stress were rescued BIP treatment. Although BIP treatment decreased some iron accumulation in the cytoplasm and mitochondria, the accumulation of iron in the lysosomes and levels of cellular ROS were unaffected. In addition, the change of specific RNA levels related to free radical stress in patient fibroblasts was not rescued by BIP. To alleviate free radical stress, we investigated whether l-serine can regulate abnormal structures in cells derived from patients with BPAN through the regulation of free radical stress. l-serine treatment alleviated increase of enlarged lysosomes and iron accumulation and rescued impaired lysosomal activity by reducing oxidized lipid accumulation in the lysosomes of the cells. Lamellated lipids in the lysosomes of the cells were identified as lipofuscin through correlative light and electron microscopy, and l-serine treatment reduced the increase of lipofuscin. These data suggest that l-serine reduces oxidative stress-mediated lysosomal lipid oxidation and iron accumulation by rescuing lysosomal activity.

Automated Differentiation of Atypical Parkinsonian Syndromes Using Brain Iron Patterns in Susceptibility Weighted Imaging.

In recent studies, iron overload has been reported in atypical parkinsonian syndromes. The topographic patterns of iron distribution in deep brain nuclei vary by each subtype of parkinsonian syndrome, which is affected by underlying disease pathologies. In this study, we developed a novel framework that automatically analyzes the disease-specific patterns of iron accumulation using susceptibility weighted imaging (SWI). We constructed various machine learning models that can classify diseases using radiomic features extracted from SWI, representing distinctive iron distribution patterns for each disorder. Since radiomic features are sensitive to the region of interest, we used a combination of T1-weighted MRI and SWI to improve the segmentation of deep brain nuclei. Radiomics was applied to SWI from 34 patients with a parkinsonian variant of multiple system atrophy, 21 patients with cerebellar variant multiple system atrophy, 17 patients with progressive supranuclear palsy, and 56 patients with Parkinson's disease. The machine learning classifiers that learn the radiomic features extracted from iron-reflected segmentation results produced an average area under receiver operating characteristic curve (AUC) of 0.8607 on the training data and 0.8489 on the testing data, which is superior to the conventional classifier with segmentation using only T1-weighted images. Our radiomic model based on the hybrid images is a promising tool for automatically differentiating atypical parkinsonian syndromes.

Atypical Young-onset Dementia in Cerebral Thromboangiitis Obliterans: A Case Report.

Young-onset dementia (YOD, age at onset below 45 y) has a broad differential diagnosis. We describe a 41-year-old man with atypical manifestations of YOD syndrome in cerebral thromoboangiitis obliterans (CTAO). Extensive antemortem workup including clinical assessment, laboratory investigations, neuroimaging, and genetic testing did not elucidate a diagnosis. Postmortem neuropathologic examination revealed cortical sickle-shaped granular atrophy, resulting from numerous remote infarcts and cortical microinfarcts that mainly affected the bilateral frontal and parietal lobe, confirming CTAO. Although CTAO is a rare cause of vascular dementia, it should be considered as one of the differentials in patients with YOD with a history of heavy smoking and presence of symmetric damages of watershed-territory on neuroimaging.

Clinical, electrophysiological, and genetic characteristics of cerebrotendinous xanthomatosis in South Korea.

Cerebrotendinous xanthomatosis (CTX) is an autosomal recessive lipid storage disorder caused by 27-hydroxylase deficiency. We report the clinical characteristics of six Korean CTX patients. The median age of onset was 22.5 years, the median age at diagnosis was 42 years, and the diagnostic delay was 18.1 years. The most common clinical symptoms were tendon xanthoma and spastic paraplegia. Four of five patients exhibited latent central conduction dysfunction. All patients carried the same mutation in CYP27A1 (c.1214 G>A [p.R405Q]). CTX is a treatable neurodegenerative disorder; however, our results revealed that patients with CTX in Korea might receive the diagnosis after a prolonged delay. .

Iron Accumulation and Changes in Cellular Organelles in WDR45 Mutant Fibroblasts.

Iron overload in the brain, defined as excess stores of iron, is known to be associated with neurological disorders. In neurodegeneration accompanied by brain iron accumulation, we reported a specific point mutation, c.974-1G>A in WD Repeat Domain 45 (WDR45), showing iron accumulation in the brain, and autophagy defects in the fibroblasts. In this study, we investigated whether fibroblasts with mutated WDR45 accumulated iron, and other effects on cellular organelles. We first identified the main location of iron accumulation in the mutant fibroblasts and then investigated the effects of this accumulation on cellular organelles, including lipid droplets, mitochondria and lysosomes. Ultrastructure analysis using transmission electron microscopy (TEM) and confocal microscopy showed structural changes in the organelles. Increased numbers of lipid droplets, fragmented mitochondria and increased numbers of lysosomal vesicles with functional disorder due to WDR45 deficiency were observed. Based on correlative light and electron microscopy (CLEM) findings, most of the iron accumulation was noted in the lysosomal vesicles. These changes were associated with defects in autophagy and defective protein and organelle turnover. Gene expression profiling analysis also showed remarkable changes in lipid metabolism, mitochondrial function, and autophagy-related genes. These data suggested that functional and structural changes resulted in impaired lipid metabolism, mitochondrial disorder, and unbalanced autophagy fluxes, caused by iron overload.

Differential Effect of Iron and Myelin on Susceptibility MRI in the Substantia Nigra.

Background The heterogeneous composition of substantia nigra (SN), including iron, nigrosome-1 substructure, and myelinated white matter, complicates the interpretation of MRI signals. Purpose To investigate R2* and quantitative susceptibility mapping (QSM) in the SN subdivisions of participants with Parkinson disease and healthy control subjects. Materials and Methods In this prospective study conducted from November 2018 to November 2019, participants with Parkinson disease and sex-matched healthy control subjects underwent 3-T MRI. R2* and QSM values were measured and compared in the anterior SN and posterior SN at the rostral (superior) and caudal (inferior) levels. Postmortem MRI and histology correlation of midbrain tissues was evaluated to investigate the effect of myelin and iron in the SN on R2* and QSM values. Results Forty individuals were evaluated: 20 healthy control subjects (mean age, 61 years ± 3 [standard deviation]; 10 men) and 20 participants with Parkinson disease (mean age, 61 years ± 4; 10 men). The R2* values of participants with Parkinson disease were higher in all subdivisions of the SN compared with R2* values in healthy control subjects (all P < .05). For QSM, no evidence of a difference was found in the rostral posterior SN (healthy control subjects, 54.1 ppb ± 21.0; Parkinson disease, 62.2 ppb ± 19.8; P = .49). The combination of rostral R2* and caudal QSM values resulted in an area under the receiver operating characteristic curve of 0.84. R2* values showed higher correlation with QSM values at the caudal level than at the rostral level within each group (all P < .001). Postmortem investigation demonstrated that R2* and QSM values showed weak correlation in the myelin-rich areas (r = 0.22 and r = 0.36, P < .001) and strong correlation in myelin-scanty areas (r ranged from approximately 0.52 to approximately 0.78, P < .001) in the SN. Conclusion Considering the iron and myelin distribution in the substantia nigra subdivisions, the subdivisional analysis of substantia nigra using R2* and quantitative susceptibility mapping might aid in specifically differentiating individuals with Parkinson disease from healthy control subjects. © RSNA, 2021 Online supplemental material is available for this article.

Impact of image compression on deep learning-based mammogram classification.

Image compression is used in several clinical organizations to help address the overhead associated with medical imaging. These methods reduce file size by using a compact representation of the original image. This study aimed to analyze the impact of image compression on the performance of deep learning-based models in classifying mammograms as "malignant"-cases that lead to a cancer diagnosis and treatment-or "normal" and "benign," non-malignant cases that do not require immediate medical intervention. In this retrospective study, 9111 unique mammograms-5672 normal, 1686 benign, and 1754 malignant cases were collected from the National Cancer Center in the Republic of Korea. Image compression was applied to mammograms with compression ratios (CRs) ranging from 15 to 11 K. Convolutional neural networks (CNNs) with three convolutional layers and three fully-connected layers were trained using these images to classify a mammogram as malignant or not malignant across a range of CRs using five-fold cross-validation. Models trained on images with maximum CRs of 5 K had an average area under the receiver operating characteristic curve (AUROC) of 0.87 and area under the precision-recall curve (AUPRC) of 0.75 across the five folds and compression ratios. For images compressed with CRs of 10 K and 11 K, model performance decreased (average 0.79 in AUROC and 0.49 in AUPRC). Upon generating saliency maps that visualize the areas each model views as significant for prediction, models trained on less compressed (CR < = 5 K) images had maps encapsulating a radiologist's label, while models trained on images with higher amounts of compression had maps that missed the ground truth completely. In addition, base ResNet18 models pre-trained on ImageNet and trained using compressed mammograms did not show performance improvements over our CNN model, with AUROC and AUPRC values ranging from 0.77 to 0.87 and 0.52 to 0.71 respectively when trained and tested on images with maximum CRs of 5 K. This paper finds that while training models on images with increased the robustness of the models when tested on compressed data, moderate image compression did not substantially impact the classification performance of DL-based models.

Iron accumulation in the oculomotor nerve of the progressive supranuclear palsy brain.

Abnormal iron accumulation around the substantia nigra (SN) is a diagnostic indicator of Parkinsonism. This study aimed to identify iron-related microarchitectural changes around the SN of brains with progressive supranuclear palsy (PSP) via postmortem validations and in vivo magnetic resonance imaging (MRI). 7 T high-resolution MRI was applied to two postmortem brain tissues, from one normal brain and one PSP brain. Histopathological examinations were performed to demonstrate the molecular origin of the high-resolution postmortem MRI findings, by using ferric iron staining, myelin staining, and two-dimensional laser ablation-inductively coupled plasma-mass spectrometry (LA-ICP-MS) imaging. In vivo iron-related MRI was performed on five healthy controls, five patients with Parkinson's disease (PD), and five patients with PSP. In the postmortem examination, excessive iron deposition along the myelinated fiber at the anterior SN and third cranial nerve (oculomotor nerve) fascicles of the PSP brain was verified by LA-ICP-MS. This region corresponded to those with high R2* values and positive susceptibility from quantitative susceptibility mapping (QSM), but was less sensitive in Perls' Prussian blue staining. In in vivo susceptibility-weighted imaging, hypointense pixels were observed in the region between the SN and red nucleus (RN) in patients with PSP, but not in healthy controls and patients with PD. R2* and QSM values of such region were significantly higher in patients with PSP compared to those in healthy controls and patients with PD as well (vs. healthy control: p = 0.008; vs. PD: p = 0.008). Thus, excessive iron accumulation along the myelinated fibers at the anterior SN and oculomotor nerve fascicles may be a pathological characteristic and crucial MR biomarker in a brain with PSP.

Brain MRI Pattern Recognition in Neurodegeneration With Brain Iron Accumulation.

Most neurodegeneration with brain iron accumulation (NBIA) disorders can be distinguished by identifying characteristic changes on magnetic resonance imaging (MRI) in combination with clinical findings. However, a significant number of patients with an NBIA disorder confirmed by genetic testing have MRI features that are atypical for their specific disease. The appearance of specific MRI patterns depends on the stage of the disease and the patient's age at evaluation. MRI interpretation can be challenging because of heterogeneously acquired MRI datasets, individual interpreter bias, and lack of quantitative data. Therefore, optimal acquisition and interpretation of MRI data are needed to better define MRI phenotypes in NBIA disorders. The stepwise approach outlined here may help to identify NBIA disorders and delineate the natural course of MRI-identified changes.

Quantitative Validation of a Visual Rating Scale for Defining High-Iron Putamen in Patients With Multiple System Atrophy.

Objectives: To validate a visual rating scale reflecting sub-regional patterns of putaminal hypointensity in susceptibility-weighted imaging of patients with multiple system atrophy (MSA). Methods: Using a visual rating scale (from G0 to G3), 2 examiners independently rated putaminal hypointensities of 37 MSA patients and 21 control subjects. To investigate the correlation with the scales, R2* values and the volume of the entire putamen were measured. Results: MSA patients with parkinsonian variant had significantly higher scores than those with cerebellar variant. Visual rating scores in MSA were correlated with R2* values [General estimating equation (GEE), Wald chi-square = 25.89, corrected p < 0.001] and volume (Wald chi-square = 75.44, corrected p < 0.001). They correlated with UPDRS motor scores. Binary logistic regression analyses revealed that the visual rating scale was a significant predictor for discriminating MSA patients from controls [multivariate model adjusted for age and sex, odds ratio 52.722 (corrected p = 0.009)]. Pairwise comparison between areas under the curve (AUCs) revealed that the visual rating scale demonstrated higher accuracy than R2* values [difference between AUCs; univariate model = 0.247 (corrected p < 0.001); multivariate model = 0.186 (corrected p = 0.003)]. There were no significant differences in clinical characteristics between the high-iron group, defined as putamen with visual rating scale ≥ G2 and R2* values ≥ third quartile, and the remaining patients. Conclusion: The visual rating scale, which reflects quantitative iron content and atrophy of the putamen as well as motor severities, could be useful for the discrimination and evaluation of patients with MSA.

Anti-CV2/CRMP5 Paraneoplastic Chorea Effectively Managed with Intravenous Amantadine.

Background: Paraneoplastic chorea is typically a subacute progressive hyperkinetic movement disorder. The mainstay of treatment is managing the underlying neoplasm. However, the clinical course may be variable, and effective symptomatic management can precede the start of cancer treatment. Case report: A 63-year-old man presented with insidious onset, slowly progressive generalized chorea for 1 year, later diagnosed as anti-CV2/CRMP5 autoantibody positive paraneoplastic chorea. His chorea was markedly improved with intravenous amantadine. Discussion: In patients with anti-CV2/CRMP5 autoantibody-related chorea, sequential follow-up of brain magnetic resonance imaging reveals progression from active inflammation to atrophy. Our report highlights the efficacy of intravenous amantadine in paraneoplastic chorea.

Acidocalcisomes: Ultrastructure, Biogenesis, and Distribution in Microbial Eukaryotes.

Acidocalcisomes are membrane-enclosed organelles with acidic lumens that accumulate polyphosphate, often in granular form, and sequester calcium and metals. They carry a transmembrane polyphosphate polymerase and two classes of proton pumps: H+-pyrophosphatases (H+-PPases) and V-type ATPases. This report describes acidocalcisomes that were snap-frozen in living cells, primarily the green alga Chlamydomonas reinhardtii, and then fractured and etched (QFDEEM). Polyphosphate granules prove to be uncommon in log-phase C. reinhardtii cells and abundant in stressed cells, where they are also found within autophagy-related vacuoles. Their E (ectoplasmic) fracture face adopts a unique rugose morphology with etching, and displays ∼14nm globular domains in broken cell preparations. Using etched membrane morphology as a guide, acidocalcisomes were identified during assembly in the trans-Golgi and were recognized in QFDEEM replicas of 18 additional algae and protists. Phylogenetic analysis documents that the eukaryotic gene encoding the signature acidocalcisomal H+-PPase pump has homologues in three widespread eukaryotic clades and has been lost in opisthokonts and Amoebozoa. The eukaryotic clades are related to three functionally diverged prokaryotic PPase pumps, one of which transports Na+. Our data indicate that the Last Eukaryotic Common Ancestor (LECA) encoded two bacteria-derived pumps and one Asgard-archaea-derived pump.

Speculating the timing of iron deposition in the putamen in multiple system atrophy.

BACKGROUND & OBJECTIVE: Although iron accumulation is thought to be associated with neurodegenerative processes, the timing of putaminal iron deposition during the disease course of multiple system atrophy (MSA) remains unclear. We sought to investigate the temporal pattern of iron deposition in the putamen of MSA patients by calculating the conditional probabilities (CPs) of multimodal MRI changes. METHODS: We simultaneously measured putaminal R2*, volume and MD values of 39 probable MSA patients and 22 control subjects. The presence of significant MRI changes was defined as higher R2* and MD values, or lower volumes than cut-off values derived from mean values in control putamen. The CPs of R2* changes without MD or volume changes were then compared with those of MD or volume changes without R2* changes. RESULTS: Regardless of the cut-off values, the CPs of R2* increments without MD or volume changes were significantly lower than those for MD or volume changes without R2* increments. The associations of R2* with volume and MD values appeared in non-linear exponential and quadratic patterns, respectively. CONCLUSIONS: Our findings suggest that putaminal iron accumulation would occur under volume atrophy or MD increments. Thus, iron deposition in the putamen of MSA patients is likely a secondary byproduct of neurodegeneration.

Brain Iron Accumulation in Atypical Parkinsonian Syndromes: in vivo MRI Evidences for Distinctive Patterns.

Recent data suggest mechanistic links among perturbed iron homeostasis, oxidative stress, and misfolded protein aggregation in neurodegenerative diseases. Iron overload and toxicity toward dopaminergic neurons have been established as playing a role in the pathogenesis of Parkinson's disease (PD). Brain iron accumulation has also been documented in atypical parkinsonian syndromes (APS), mainly comprising multiple system atrophy (MSA), and progressive supranuclear palsy (PSP). Iron-sensitive magnetic resonance imaging (MRI) has been applied to identify iron-related signal changes for the diagnosis and differentiation of these disorders. Topographic patterns of widespread iron deposition in deep brain nuclei have been described as differing between patients with MSA and PSP and those with PD. A disease-specific increase of iron occurs in the brain regions mainly affected by underlying disease pathologies. However, whether iron changes are a primary pathogenic factor or an epiphenomenon of neuronal degeneration has not been fully elucidated. Moreover, the clinical implications of iron-related pathology in APS remain unclear. In this review study, we collected data from qualitative and quantitative MRI studies on brain iron accumulation in APS to identify disease-related patterns and the potential role of iron-sensitive MRI.

Visually interpretable deep network for diagnosis of breast masses on mammograms.

Recently, deep learning technology has achieved various successes in medical image analysis studies including computer-aided diagnosis (CADx). However, current CADx approaches based on deep learning have a limitation in interpreting diagnostic decisions. The limited interpretability is a major challenge for practical use of current deep learning approaches. In this paper, a novel visually interpretable deep network framework is proposed to provide diagnostic decisions with visual interpretation. The proposed method is motivated by the fact that the radiologists characterize breast masses according to the breast imaging reporting and data system (BIRADS). The proposed deep network framework consists of a BIRADS guided diagnosis network and a BIRADS critic network. A 2D map, named BIRADS guide map, is generated in the inference process of the deep network. The visual features extracted from the breast masses could be refined by the BIRADS guide map, which helps the deep network to focus on more informative areas. The BIRADS critic network makes the BIRADS guide map to be relevant to the characterization of masses in terms of BIRADS description. To verify the proposed method, comparative experiments have been conducted on public mammogram database. On the independent test set (170 malignant masses and 170 benign masses), the proposed method was found to have significantly higher performance compared to the deep network approach without using the BIRADS guide map (p < 0.05). Moreover, the visualization was conducted to show the location where the deep network exploited more information. This study demonstrated that the proposed visually interpretable CADx framework could be a promising approach for visually interpreting the diagnostic decision of the deep network.

Synthesis and Biological Activity of Tetrameric Ribitol Phosphate Fragments of Staphylococcus aureus Wall Teichoic Acid.

A systematically designed and synthesized ribitol phosphate (RboP) oligomer using a series of building blocks, which make up the wall teichoic acid (WTA) of S. aureus, is presented. Based on the use of a solution-phase phosphodiester synthesis, a library of ribitol phosphate tetramers, decorated with d-alanine and N-acetylglucosamine (GlcNAc), were generated. The synthesized RboP tetramers showed increased cytokine levels in mice in a subcutaneous air pouch model.

Quasi-Stem Cells Derived from Human Somatic Cells by Chemically Modified Carbon Nanotubes.

Surface modification of micro- and nanotopography was employed to alter the surface properties of scaffolds for controlling cell attachment, proliferation, and differentiation. This study reports a method for generating multinucleated colonies as evidenced by spherical colony formation through nanotopography-induced expression of reprogramming factors in human dermal fibroblasts. Colony formation was achieved by subjecting the cells to specific environments such as culturing with single-walled carbon nanotubes and poly-l-lysine (PLL-SWCNTs). We obtained encouraging results showing that PLL-SWCNT treatment transformed fibroblast cells, and the transformed cells expressed the pluripotency-associated factors OCT4, NANOG, and SOX2 in addition to TRA-1-60 and SSEA-4, which are characteristic stem cell markers. Downregulation of lamin A/C, HDAC1, HDAC6, Bcl2, cytochrome c, p-FAK, p-ERK, and p-JNK and upregulation of H3K4me3 and p-p38 were confirmed in the generated colonies, indicating reprogramming of cells. This protocol increases the possibility of successfully reprogramming somatic cells into induced pluripotent stem cells (iPSCs), thereby overcoming the difficulties in iPSC generation such as genetic mutations, carcinogenesis, and undetermined risk factors.

Specific visualization of neuromelanin-iron complex and ferric iron in the human post-mortem substantia nigra using MR relaxometry at 7T.

Neuromelanin (NM) is an endogenous iron chelating molecule of pigmented neurons in the human substantia nigra (SN). Along with the increase in iron deposition, the reduction in NM-containing dopaminergic neurons and the variation of iron load on NM are generally considered to be important factors participating to pathogenesis of Parkinson's disease (PD). The aim of this study was to non-invasively delineate the spatial distributions of paramagnetic magnetic susceptibility perturbers, such as NM-iron complex and ferric iron in SN. Multiple quantitative MR parameters of T1, T2, T2*, susceptibility weighted image (SWI), quantitative susceptibility map (QSM), and T1 weighted image with magnetization transfer (MT) effects were acquired for six post-mortem SN samples without a history of neurological disease. Co-registered quantitative histological validations were performed to identify and correlate NM pigments, iron deposits, and myelin distributions with respect to associated MR parameters. The regions with NM pigments and iron deposits showed positive magnetic susceptibility (paramagnetic) values, while myelinated areas showed negative magnetic susceptibility (diamagnetic) values from the QSM. The region of reduced T2 values in SN mostly coincided with high iron deposits, but not necessarily with the NM pigments. The correlations between T2*/T2 (or T2*/T22) values and NM pigments were higher than those between T2* values and NM pigments, due to the effective size differences between NM-iron complex and ferric iron. Consequently, separate segmentations of ferric iron from the T2 map and NM-iron complex from the T2*/T2 map (or T2*/T22 map) were possible with the boundary of the SN determined from the T1 weighted image.