Displaying publications 41 - 60 of 99 in total

Abstract:
Sort:
  1. Shanmugasiva VV, Ramli Hamid MT, Fadzli F, Shaleen Kaur KS, Abd Rahman N, Rahmat K
    Malays J Pathol, 2018 Dec;40(3):349-353.
    PMID: 30580368
    INTRODUCTION: Myofibroblastoma is a rare benign mesenchymal tumour arising from the stromal elements of the breast tissue. Histopathological variants such as classic, cellular, collagenous / fibrous, lipomatous, infiltrative, myxoid and epithelioid have been identified. Most myofibroblastomas are immunoreactive for CD34, actin, CD10 and desmin, usually express oestrogen receptor (ER), progesterone receptor (PR) and variably express androgen receptor (AR).

    CASE REPORT: We report a case of myofibroblastoma in an octogenarian male presenting with painless solitary breast lump. Mammography (digital tomosynthesis) and ultrasound showed a well-circumscribed hyperdense mass and hypoechoic, solid, oval mass with peripheral vascularity respectively. Patient underwent wide local excision.

    DISCUSSION: Diverse characteristics of myofibroblastoma on imaging necessitates histopathological analysis for an accurate diagnosis. Myofibroblastoma are often confused with fibroadenomas due to the benign imaging characteristics and with malignant neoplasia due to their wide morphological spectrum. Surgical excision is considered curative.

  2. Leong WL, Lai LL, Nik Mustapha NR, Vijayananthan A, Rahmat K, Mahadeva S, et al.
    J Gastroenterol Hepatol, 2020 Jan;35(1):135-141.
    PMID: 31310032 DOI: 10.1111/jgh.14782
    BACKGROUND AND AIM: Transient elastography (TE) and point shear wave elastography (pSWE) are noninvasive methods to diagnose fibrosis stage in patients with chronic liver disease. The aim of this study is to compare the accuracy of the two methods to diagnose fibrosis stage in non-alcoholic fatty liver disease (NAFLD) and to study the intra-observer and inter-observer variability when the examinations were performed by healthcare personnel of different backgrounds.

    METHODS: Consecutive NAFLD patients who underwent liver biopsy were enrolled in this study and had two sets each of pSWE and TE examinations by a nurse and a doctor on the same day of liver biopsy procedure. The medians of the four sets of pSWE and TE were used for evaluation of diagnostic accuracy using area under receiver operating characteristic curve (AUROC). Intra-observer and inter-observer variability was analyzed using intraclass correlation coefficients.

    RESULTS: Data for 100 NAFLD patients (mean age 57.1 ± 10.2 years; male 46.0%) were analyzed. The AUROC of TE for diagnosis of fibrosis stage ≥ F1, ≥ F2, ≥ F3, and F4 was 0.89, 0.83, 0.83, and 0.89, respectively. The corresponding AUROC of pSWE was 0.80, 0.72, 0.69, and 0.79, respectively. TE was significantly better than pSWE for the diagnosis of fibrosis stages ≥ F2 and ≥ F3. The intra-observer and inter-observer variability of TE and pSWE measurements by the nurse and doctor was excellent with intraclass correlation coefficient > 0.96.

    CONCLUSION: Transient elastography was significantly better than pSWE for the diagnosis of fibrosis stage ≥ F2 and ≥ F3. Both TE and pSWE had excellent intra-observer and inter-observer variability when performed by healthcare personnel of different backgrounds.

  3. Sanmugasiva VV, Ramli Hamid MT, Fadzli F, Rozalli FI, Yeong CH, Ab Mumin N, et al.
    Sci Rep, 2020 11 26;10(1):20628.
    PMID: 33244075 DOI: 10.1038/s41598-020-77456-6
    This study aims to assess the diagnostic accuracy of digital breast tomosynthesis in combination with full field digital mammography (DBT + FFDM) in the charaterisation of Breast Imaging-reporting and Data System (BI-RADS) category 3, 4 and 5 lesions. Retrospective cross-sectional study of 390 patients with BI-RADS 3, 4 and 5 mammography with available histopathology examination results were recruited from in a single center of a multi-ethnic Asian population. 2 readers independently reported the FFDM and DBT images and classified lesions detected (mass, calcifications, asymmetric density and architectural distortion) based on American College of Radiology-BI-RADS lexicon. Of the 390 patients recruited, 182 malignancies were reported. Positive predictive value (PPV) of cancer was 46.7%. The PPV in BI-RADS 4a, 4b, 4c and 5 were 6.0%, 38.3%, 68.9%, and 93.1%, respectively. Among all the cancers, 76% presented as masses, 4% as calcifications and 20% as asymmetry. An additional of 4% of cancers were detected on ultrasound. The sensitivity, specificity, PPV and NPV of mass lesions detected on DBT + FFDM were 93.8%, 85.1%, 88.8% and 91.5%, respectively. The PPV for calcification is 61.6% and asymmetry is 60.7%. 81.6% of cancer detected were invasive and 13.3% were in-situ type. Our study showed that DBT is proven to be an effective tool in the diagnosis and characterization of breast lesions and supports the current body of literature that states that integrating DBT to FFDM allows good characterization of breast lesions and accurate diagnosis of cancer.
  4. Mumin NA, Rahmat K, Fadzli F, Ramli MT, Westerhout CJ, Ramli N, et al.
    Sci Rep, 2019 02 06;9(1):1459.
    PMID: 30728394 DOI: 10.1038/s41598-018-37451-4
    Synthesized 2D images can be reconstructed from tomosynthesis images in breast imaging. This study aims to investigate the diagnostic efficacy of synthesized 2D images (C-View) in comparison to full field digital mammography (FFDM) when used with digital breast tomosynthesis (DBT) in multi-ethnic Malaysian population. FFDM and C-View images (n = 380) were independently evaluated by three readers through Breast Imaging Reporting and Data System (BI-RADS) categorisation, breast density and lesion characterisation. Statistical analysis was done comparing sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of C-View + DBT with FFDM + DBT as standard of reference. Very good interreader agreement in BI-RADS category and density assessment between C-View + DBT and FFDM + DBT, with high sensitivity, specificity, PPV and NPV of C-View + DBT when compared with FFDM + DBT. There was comparable PPV between C-View + DBT and FFDM + DBT, with histopathology as gold standard. High level of interreader agreement in BI-RADS category and density assessment for FFDM + DBT and C-View + DBT. There was good agreement between FFDM + DBT and C-View + DBT in mass characterization, and almost perfect agreement in calcification and asymmetric density. 52.2% lower radiation dose incurred when using C-View + DBT. Hence, synthesized 2D images are comparable to FFDM with reduction in radiation dose within the limits of Malaysian multi-ethnic population.
  5. Tay CG, Ariffin H, Yap S, Rahmat K, Sthaneshwar P, Ong LC
    J Child Neurol, 2015 Jun;30(7):927-31.
    PMID: 25122112 DOI: 10.1177/0883073814540523
    Succinic semialdehyde dehydrogenase deficiency is a rare autosomal recessive disorder affecting catabolism of the neurotransmitter gamma-aminobutyric acid (GABA), with a wide range of clinical phenotype. We report a Malaysian Chinese boy with a severe early onset phenotype due to a previously unreported mutation. Urine organic acid chromatogram revealed elevated 4-hydroxybutyric acid. Magnetic resonance imaging (MRI) of the brain demonstrated cerebral atrophy with atypical putaminal involvement. Molecular genetic analysis showed a novel homozygous 3-bp deletion at the ALDH5A1 gene c.1501_1503del (p.Glu501del). Both parents were confirmed to be heterozygotes for the p.Glu501del mutation. The clinical course was complicated by the development of subdural hemorrhage probably as a result of rocking the child to sleep for erratic sleep-wake cycles. This case illustrates the need to recognize that trivial or unintentional shaking of such children, especially in the presence of cerebral atrophy, can lead to subdural hemorrhage.
  6. Chan WY, Hamid MTR, Gowdh NFM, Rahmat K, Yaakup NA, Chai CS
    Curr Med Imaging, 2021;17(6):677-685.
    PMID: 33390122 DOI: 10.2174/1573405616666201231103312
    BACKGROUND: Coronavirus disease 2019 (COVID-19) is highly contagious and has claimed more than one million lives, besides causing hardship and disruptions. The Fleischner Society has recommended chest X-ray (CXR) in detecting cases at high risk of disease progression, for triaging suspected patients with moderate-to-severe illness, and for eliminating false negatives in areas with high pre-test probability or limited resources. Although CXR is less sensitive than real-- time reverse transcription-polymerase chain reaction (RT-PCR) in detecting mild COVID-19, it is nevertheless useful because of equipment portability, low cost and practicality in serial assessments of disease progression among hospitalized patients.

    OBJECTIVE: This study aims to review the typical and relatively atypical CXR manifestations of COVID-19 pneumonia in a tertiary care hospital.

    METHODS: The CXRs of 136 COVID-19 patients confirmed through real-time RT-PCR from March to May 2020 were reviewed. A literature search was performed using PubMed.

    RESULTS: A total of 54 patients had abnormal CXR whilst the others were normal. Typical CXR findings included pulmonary consolidation or ground-glass opacities in a multifocal, bilateral peripheral, or lower zone distribution, whereas atypical CXR features comprised cavitation and pleural effusion.

    CONCLUSION: Typical findings of COVID-19 infection in chest computed tomography studies can also be seen in CXR. The presence of atypical features associated with worse disease outcome. Recognition of these features on CXR will improve the accuracy and speed of diagnosing COVID-19 patients.

  7. Ramli NM, Sidek S, Rahman FA, Peyman M, Zahari M, Rahmat K, et al.
    Graefes Arch Clin Exp Ophthalmol, 2014 Jun;252(6):995-1000.
    PMID: 24770532 DOI: 10.1007/s00417-014-2622-6
    PURPOSE: To measure optic nerve (ON) volume using 3 T magnetic resonance imaging (MRI), to correlate ON volume with retinal nerve fiber layer (RNFL) thickness, and to determine the viability of MRI as an objective tool in distinguishing glaucoma severity.

    METHODS: In this cross-sectional study, 30 severe glaucoma patients, 30 mild glaucoma patients and 30 age-matched controls were recruited. All subjects underwent standard automated perimetry, RNFL analysis and 3 T MRI examinations. Glaucoma patients were classified according to the Hodapp-Anderson-Parish classification. Pearson's correlation coefficient was used to correlate ON volume with RNFL, and receiver operating curve (ROC) analysis was performed to determine the sensitivity and specificity of ON volume in detecting glaucoma severity.

    RESULTS: Optic nerve volume was significantly lower in both the left and right eyes of the severe glaucoma group (168.70 ± 46.28 mm(3); 167.40 ± 45.36 mm(3)) than in the mild glaucoma group (264.03 ± 78.53 mm(3); 264.76 ± 78.88 mm(3)) and the control group (297.80 ± 71.45 mm(3); 296.56 ± 71.02 mm(3)). Moderate correlation was observed between: RNFL thickness and ON volume (r = 0.51, p <0.001), and in mean deviation of visual field and optic nerve volume (r = 0.60, p 

  8. Tan M, Al-Shabi M, Chan WY, Thomas L, Rahmat K, Ng KH
    Med Biol Eng Comput, 2021 Feb;59(2):355-367.
    PMID: 33447988 DOI: 10.1007/s11517-021-02313-1
    This study objectively evaluates the similarity between standard full-field digital mammograms and two-dimensional synthesized digital mammograms (2DSM) in a cohort of women undergoing mammography. Under an institutional review board-approved data collection protocol, we retrospectively analyzed 407 women with digital breast tomosynthesis (DBT) and full-field digital mammography (FFDM) examinations performed from September 1, 2014, through February 29, 2016. Both FFDM and 2DSM images were used for the analysis, and 3216 available craniocaudal (CC) and mediolateral oblique (MLO) view mammograms altogether were included in the dataset. We analyzed the mammograms using a fully automated algorithm that computes 152 structural similarity, texture, and mammographic density-based features. We trained and developed two different global mammographic image feature analysis-based breast cancer detection schemes for 2DSM and FFDM images, respectively. The highest structural similarity features were obtained on the coarse Weber Local Descriptor differential excitation texture feature component computed on the CC view images (0.8770) and MLO view images (0.8889). Although the coarse structures are similar, the global mammographic image feature-based cancer detection scheme trained on 2DSM images outperformed the corresponding scheme trained on FFDM images, with area under a receiver operating characteristic curve (AUC) = 0.878 ± 0.034 and 0.756 ± 0.052, respectively. Consequently, further investigation is required to examine whether DBT can replace FFDM as a standalone technique, especially for the development of automated objective-based methods.
  9. Sanmugasiva V, Hamid MTR, Fadzli F, Ab Mumin N, Rahmat K
    Curr Med Imaging, 2021 Oct 04.
    PMID: 34607549 DOI: 10.2174/1573405617666211004114041
    INTRODUCTION: Metaplastic breast carcinoma is an uncommon malignancy that constitutes < 5% of all breast cancers. There are 5 subtypes which are spindle cell, squamous cell, carcinosarcoma, matrix-producing and metaplastic with osteoclastic giant cells. Spindle cell carcinoma represents approximately <0.3% of invasive breast carcinomas. It is typically a triple-negative cancer with distinct pathological characteristics, but relatively a non-conclusive imaging findings.

    CASE REPORT: An elderly lady presented with an enlarging painful left breast lump for 1 year. Palpable left breast lump noted on clinical examination. Mammography demonstrated a high density, oval lesion with a partially indistinct margin. Corresponding ultrasound showed a large irregular heterogeneous lesion with solid-cystic areas. Histopathology showed atypical spindle-shaped cells which stained positive for cytokeratins and negative for hormone and human epidermal growth factor receptors, which favours spindle cell metaplastic carcinoma. Left mastectomy and axillary dissection were performed, and the final diagnosis was consistent with metaplastic spindle cell carcinoma.

    CONCLUSION: Spindle cell carcinoma of the breast is a rare aggressive histological type of carcinoma which may present with benign features on imaging. Tissue diagnosis is essential for prompt diagnosis with multidisciplinary team discussion to guide management and improve patient's outcome.

  10. Mohd Nor NS, Fong CY, Rahmat K, Vanessa Lee WM, Zaini AA, Jalaludin MY
    Eur Endocrinol, 2018 Apr;14(1):59-61.
    PMID: 29922355 DOI: 10.17925/EE.2018.14.1.59
    Cerebral oedema is the most common neurological complication of diabetic ketoacidosis (DKA). However, ischaemic and haemorrhagic brain injury has been reported infrequently. A 10-year old girl who was previously well presented with severe DKA. She was tachycardic with poor peripheral perfusion but normotensive. However, two fast boluses totalling 40 ml/kg normal saline were given. She was transferred to another hospital where she was intubated due to drowsiness. Rehydration fluid (maintenance and 48-hour correction for 7.5% dehydration) was started followed by insulin infusion. She was extubated within 24 hours of admission. Her ketosis resolved soon after and subcutaneous insulin was started. However, about 48 hours after admission, her Glasgow Coma Scale score dropped to 11/15 (E4M5V2) with expressive aphasia and upper motor neuron signs. One dose of mannitol was given. Her symptoms improved gradually and at 26-month follow-up she had a near-complete recovery with only minimal left lower limb weakness. Serial magnetic resonance imaging brain scans showed vascular ischaemic injury at the frontal-parietal watershed regions with haemorrhagic transformation. This case reiterates the importance of monitoring the neurological status of patient's with DKA closely for possible neurological complications including an ischaemic and haemorrhagic stroke.
  11. Fong CY, Khine MM, Peter AB, Lim WK, Rozalli FI, Rahmat K
    J Clin Neurosci, 2017 Feb;36:73-75.
    PMID: 27887978 DOI: 10.1016/j.jocn.2016.10.050
    A 14-year-old girl presented with encephalopathy, delirium and ophthalmoplegia following a 3day history of high-grade fever. Brain MRI on day 6 of illness showed diffusion restricted ovoid lesion in the splenium of corpus callosum. Dengue virus encephalitis was diagnosed with positive PCR for dengue virus type-2 in both serum and cerebrospinal fluid. She made a complete recovery from day 10 of illness. Repeat brain MRI on day 12 of illness showed resolution of the splenial lesion. Serial diffusion tensor imaging (DTI) showed normal fractional anisotropy values on resolution of splenial lesion indicating that MERS was likely due to transient interstitial oedema with preservation of white matter tracts. This is the first reported case of MERS following dengue virus infection. It highlights the usefulness of performing serial DTI in understanding the underlying pathogenesis of MERS. Our case report widens the neurological manifestations associated with dengue infection and reiterates that patients with MERS should be managed supportively as the splenial white matter tracts are reversibly involved in MERS.
  12. Liong CC, Rahmat K, Mah JS, Lim SY, Tan AH
    Can J Neurol Sci, 2016 Sep;43(5):719-20.
    PMID: 27670213 DOI: 10.1017/cjn.2016.269
  13. Ramli N, Yap A, Muridan R, Seow P, Rahmat K, Fong CY, et al.
    Clin Radiol, 2020 01;75(1):77.e15-77.e22.
    PMID: 31668796 DOI: 10.1016/j.crad.2019.09.134
    AIM: To evaluate the microstructural abnormalities of the white matter tracts (WMT) using diffusion tensor imaging (DTI) in children with global developmental delay (GDD).

    MATERIALS AND METHODS: Sixteen children with GDD underwent magnetic resonance imaging (MRI) and cross-sectional DTI. Formal developmental assessment of all GDD patients was performed using the Mullen Scales of Early Learning. An automated processing pipeline for the WMT assessment was implemented. The DTI-derived metrics of the children with GDD were compared to healthy children with normal development (ND).

    RESULTS: Only two out of the 17 WMT demonstrated significant differences (p<0.05) in DTI parameters between the GDD and ND group. In the uncinate fasciculus (UF), the GDD group had lower mean values for fractional anisotropy (FA; 0.40 versus 0.44), higher values for mean diffusivity (0.96 versus 0.91×10-3 mm2/s) and radial diffusivity (0.75 versus 0.68×10-3 mm2/s) compared to the ND group. In the superior cerebellar peduncle (SCP), mean FA values were lower for the GDD group (0.38 versus 0.40). Normal myelination pattern of DTI parameters was deviated against age for GDD group for UF and SCP.

    CONCLUSION: The UF and SCP WMT showed microstructural changes suggestive of compromised white matter maturation in children with GDD. The DTI metrics have potential as imaging markers for inadequate white matter maturation in GDD children.

  14. Ramli Hamid MT, Rahmat K, Hamid SA, Kirat Singh SK, Hooi TG
    Curr Med Imaging Rev, 2019;15(9):866-872.
    PMID: 32008533 DOI: 10.2174/1573405614666180627101520
    BACKGROUND: Breast cancer is the commonest cancer affecting Malaysian women, accounting for an estimated 30% of all new cancer diagnosed annually. Improvements in breast cancer management have increased the breast cancer survival rate in Malaysia. Clinical and radiological surveillance of the treated breast is vital, as early detection of recurrence improves patient's survival rate.

    DISCUSSION: As surgery and radiotherapy alter the appearance of the breasts, distinguishing between recurrence and benign post-surgical changes can be challenging radiologically due to overlapping features. Despite this, differentiation between these two entities is usually possible by recognizing characteristic features of post-treatment sequelae and the evolution of the appearance of the conservatively treated breast by comparing interval findings on serial studies.

    CONCLUSION: This pictorial review aims to describe the typical and unusual features of post-treated breasts in the multimodality imaging workup of an established breast care centre in a teaching hospital in Malaysia.

  15. Ng WL, Omar N, Ab Mumin N, Ramli Hamid MT, Vijayananthan A, Rahmat K
    Acad Radiol, 2022 Jan;29 Suppl 1:S69-S78.
    PMID: 33926793 DOI: 10.1016/j.acra.2021.03.018
    OBJECTIVES: This study evaluates the diagnostic performance of shear wave elastography (SWE) in differentiating between benign and axillary lymph node (ALN) metastasis in breast carcinoma.

    MATERIALS AND METHODS: Breast lesions and axillae of 107 patients were assessed using B-mode ultrasound and SWE. Histopathology was the diagnostic gold standard.

    RESULTS: In metastatic axillary lymph nodes, qualitative SWE using color patterns had the highest area under curve (AUC) value, followed by B-mode Ultrasound (cortical thickening >3 mm) and quantitative SWE using Emax of 15.2 kPa (AUC of 81.3%, 70.1%, and 61.2%, respectively). Qualitative SWE exhibited better diagnostic performance than the other two parameters, with sensitivity of 96.0% and specificity of 56.1%. Combination of B-mode Ultrasound (using cortical thickness of >3 mm as cut-off point) and qualitative SWE (Color patterns of 2 to 4) showed sensitivity of 71.6%, specificity of 95%, PPV of 96%, NPV of 66.7%, and accuracy of 80.4%.

    CONCLUSION: Qualitative SWE assessment exhibited higher accuracy compared to quantitative values. Qualitative SWE as an adjunct to B-mode ultrasound can further improve the diagnostic accuracy of metastatic ALN in breast cancer.

  16. Ab Mumin N, Ramli Hamid MT, Wong JHD, Rahmat K, Ng KH
    Acad Radiol, 2022 Jan;29 Suppl 1:S89-S106.
    PMID: 34481705 DOI: 10.1016/j.acra.2021.07.017
    OBJECTIVE: Magnetic resonance imaging (MRI) is the most sensitive imaging modality in detecting breast cancer. The purpose of this systematic review is to investigate the role of human extracted MRI phenotypes in classifying molecular subtypes of breast cancer.

    METHODS: We performed a literature search of published articles on the application of MRI phenotypic features in invasive breast cancer molecular subtype classifications by radiologists' interpretation on Medline Complete, Pubmed, and Google scholar from 1st January 2000 to 31st March 2021. Of the 1453 literature identified, 42 fulfilled the inclusion criteria.

    RESULTS: All studies were case-controlled, retrospective study and research-based. The majority of the studies assessed the MRI features using American College of Radiology- Breast Imaging Reporting and Data System (ACR-BIRADS) classification and using dynamic contrast-enhanced (DCE) kinetic features, Apparent Diffusion Coefficient (ADC) values, and T2 sequence. Most studies divided invasive breast cancer into 4 main subtypes, luminal A, luminal B, HER2, and triple-negative (TN) cancers, and used 2 readers. We present a summary of the radiologists' extracted breast MRI phenotypical features and their correlating breast cancer subtypes classifications. The characteristic features are morphology, enhancement kinetics, and T2 signal intensity. We found that the TN subtype has the most distinctive MRI features compared to the other subtypes and luminal A and B have many similar features.

    CONCLUSION: The MRI features which are predictive of each subtype are the morphology, internal enhancement features, and T2 signal intensity, predominantly between TN and the rest. Radiologists' visual interpretation of some of MRI features may offer insight into the respective invasive breast cancer molecular subtype. However, current evidence are still limited to "suggestive" features instead of a diagnostic standard.  Further research is recommended to explore this potential application, for example, by augmentation of radiologists' visual interpretation by artificial intelligence.

  17. Foo JC, Rahmat K, Mumin NA, Koh MT, Gan CS, Ramli N, et al.
    J Clin Neurosci, 2017 Sep 01.
    PMID: 28867361 DOI: 10.1016/j.jocn.2017.08.024
    We report a rare case of paediatric diphtheria complicated with encephalitis. A 6-year-old boy who did not receive his scheduled diptheria-tetanus-pertusis vaccination presented with one episode of generalised convulsive seizure. His illness was preceded by a 3day history of fever associated with enlarged exudative tonsils with a pseudomembrane. He was commenced on intravenous penicillin and oral erythromycin. However, he developed progressive encephalopathy with focal neurological deficit which required intubation on day 5 of illness. Throat swab polymerase chain reaction for diphtheria toxin A and B were positive and diphtheria antitoxin was given. Magnetic resonance imaging (MRI) of brain showed T2-weighted hyperintensities over the anterior cingulate gyri, insular cortex and cerebellum. This is the first reported MRI finding of diphtheric encephalitis. Our report highlights the importance of neuroimaging in diagnosing diphtheric encephalitis particularly in cases with unremarkable cerebrospinal findings.
  18. Chow LS, Tang GS, Solihin MI, Gowdh NM, Ramli N, Rahmat K
    SN Comput Sci, 2023;4(2):141.
    PMID: 36624807 DOI: 10.1007/s42979-022-01545-8
    Coronavirus disease 2019 (COVID-19) is a disease caused by a novel strain of coronavirus, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), severely affecting the lungs. Our study aims to combine both quantitative and qualitative analysis of the convolutional neural network (CNN) model to diagnose COVID-19 on chest X-ray (CXR) images. We investigated 18 state-of-the-art CNN models with transfer learning, which include AlexNet, DarkNet-19, DarkNet-53, DenseNet-201, GoogLeNet, Inception-ResNet-v2, Inception-v3, MobileNet-v2, NasNet-Large, NasNet-Mobile, ResNet-18, ResNet-50, ResNet-101, ShuffleNet, SqueezeNet, VGG-16, VGG-19, and Xception. Their performances were evaluated quantitatively using six assessment metrics: specificity, sensitivity, precision, negative predictive value (NPV), accuracy, and F1-score. The top four models with accuracy higher than 90% are VGG-16, ResNet-101, VGG-19, and SqueezeNet. The accuracy of these top four models is between 90.7% and 94.3%; the F1-score is between 90.8% and 94.3%. The VGG-16 scored the highest accuracy of 94.3% and F1-score of 94.3%. The majority voting with all the 18 CNN models and top 4 models produced an accuracy of 93.0% and 94.0%, respectively. The top four and bottom three models were chosen for the qualitative analysis. A gradient-weighted class activation mapping (Grad-CAM) was used to visualize the significant region of activation for the decision-making of image classification. Two certified radiologists performed blinded subjective voting on the Grad-CAM images in comparison with their diagnosis. The qualitative analysis showed that SqueezeNet is the closest model to the diagnosis of two certified radiologists. It demonstrated a competitively good accuracy of 90.7% and F1-score of 90.8% with 111 times fewer parameters and 7.7 times faster than VGG-16. Therefore, this study recommends both VGG-16 and SqueezeNet as additional tools for the diagnosis of COVID-19.
  19. Lim BK, Bux SI, Rahmat K, Lam SY, Liew YW
    Singapore Med J, 2012 Nov;53(11):732-6.
    PMID: 23192500
    We compared the effectiveness of different types of non-commercial neutral oral contrast agents for bowel distension and mural visualisation in computed tomographic (CT) enterography.
  20. Ibrahim R, Rahmat K, Fadzli F, Rozalli FI, Westerhout CJ, Alli K, et al.
    Singapore Med J, 2016 Jan 6.
    PMID: 26767894 DOI: 10.11622/smedj.2016001
    This study aimed to evaluate the vascular pattern of solid breast lesions using power Doppler ultrasound (PDUS) and to assess whether the presence of intratumoural penetrating vessels can be used to predict breast cancer malignancy.
Filters
Contact Us

Please provide feedback to Administrator ([email protected])

External Links