Search

Objective: This study aims to develop a ResNet50‐based deep learning model for focal liver lesion (FLL) classification in ultrasound images, comparing its performance with other models and prior research. Methodology: We retrospectively collected 581 ultrasound images from the Chulabhorn Hospital's HCC surveillance and screening project (2010–2018). The dataset comprised five classes: non‐FLL, hepatic cyst (Cyst), hemangioma (HMG), focal fatty sparing (FFS), and hepatocellular carcinoma (HCC). We conducted 5‐fold cross‐validation after random dataset partitioning, enhancing training data with data augmentation. Our models used modified pre‐trained ResNet50, GGN, ResNet18, and VGG16 architectures. Model performance, assessed via confusion matrices for sensitivity, specificity, and accuracy, was compared across models and with prior studies. Results: ResNet50 outperformed other models, achieving a 5‐fold cross‐validation accuracy of 87 ± 2.2%. While VGG16 showed similar performance, it exhibited higher uncertainty. In the testing phase, the pretrained ResNet50 excelled in classifying non‐FLL, cysts, and FFS. To compare with other research, ResNet50 surpassed the prior methods like two‐layered feed‐forward neural networks (FFNN) and CNN+ReLU in FLL diagnosis. Conclusion: ResNet50 exhibited good performance in FLL diagnosis, especially for HCC classification, suggesting its potential for developing computer‐aided FLL diagnosis. However, further refinement is required for HCC and HMG classification in future studies. [ABSTRACT FROM AUTHOR]

Alzheimer's disease is a significant global health challenge characterized by progressive brain degeneration associated with aging. Early detection is crucial for improved prognosis and treatment outcomes. Our study aimed to develop a support vector machine (SVM) classification model using T1-weighted MR and F-18 FDG PET brain imaging to classify cognitive normal (CN) and early-stage Alzheimer's disease, including early mild cognitive impairment (EMCI) and late mild cognitive impairment (LMCI), in individuals aged 65-75. The study comprised three steps. Firstly, image preprocessing involved uniformity processing, B1 bias field correction for MR images, and FWHM optimization with spatial/intensity normalization for PET images. Secondly, MR images were registered to the MNI305 brain template for structural segmentation, and PET images were normalized using the standardized PET template. Co-registration of normalized PET images to segmented MR images provided anatomical segmented PET uptake volumes. These served as input for the classification model. Thirdly, SVM models classified CN vs. LMCI, EMCI vs. LMCI, and CN vs. EMCI using MRI, PET, and combined PET/MR. Feature sets included all features, clinical-based features, and F1-score ranked features, resulting in 27 classification models. In CN vs. EMCI, combined PET/MR with all features achieved 0.71 AUC, 65.78% accuracy, and 68.67% specificity. For CN vs. LMCI, superior performance was observed with combined PET/MR using F1-score ranked features: 0.82 AUC, 77.78% accuracy, and 77.55% specificity. In EMCI vs. LMCI, PET alone with all features achieved the highest performance: 0.72 AUC, 69.23% accuracy, and 60.32% specificity. In conclusion, PET is pivotal in MCI stage differentiation, and using all features aids in challenging tasks (CN vs. EMCI and EMCI vs. LMCI). The combined PET/MR modality notably distinguishes CN from MCI, emphasizing the potential of multimodal imaging to enhance differentiating cognitive states in individuals. [ABSTRACT FROM AUTHOR]

Introduction: Periampullary cancer is one of the major health concerns worldwide. Even though there are some attempts to work the disease out, the prognosis of periampullary cancer remains poor. Objective: To study results and predictive factors for survival of periampullary cancer patients at Ut.aradit Hospital. Patients and Methods: A long-term observational analytic study was done. Information of 82 patients diagnosed with periampullary cancer who received treatment at Uttaradit Hospital from October 2007 to October 2012 was gathered from the database of Department of Surgery, Uttaradit Hospital. The patients' final status had been followed until September 2015. Descriptive statistics, Kaplan-Meier method and a Log-rank test were app.ied to analyze the information. Results: Patients had an average age of 64.99 years and 52.44% of them were male. All patients had primary tumors located in places as follows: 56.09% at the head of pancreas and 39.02% at the ampulla of Var.er. Pancreaticoduodenectomy (PD) was performed in 48.78% of the patients. The overall mortality rate was 53.66%. The median survival time was 2.75 years. The predictive factors were chronic alcohol drinking (p < 0.01) and PD operation (p < 0.05). Conclusion: Chronic alcohol drinking and a PD operation have significant contribution to the surv.val of periampullary cancer patients. [ABSTRACT FROM AUTHOR]

Objective. To explore prognostic characteristics for locoregional recurrence, distant recurrence, and mortality in patients with breast cancer. Methods. A 5-year retrospective review of patients was conducted in two university affiliated hospitals in the north of Thailand. Prognostic characteristics and clinical outcomes were retrieved from medical registry. Death was verified by the civil database from the Ministry of Interior, direct telephone contact, or by prepaid postcard. Data were analyzed by stratified Cox's regression proposed by Lunn & McNeil, in which multiple-typed outcomes were analyzed in a single multivariable model. Results.The assembled cohort comprised 829 patients. Under the multivariable analysis, 7 prognostic characteristics were significant prognostic indicators. Positive axillary lymph nodes >3 and presence of lymphovascular invasion (LVI) increased locoregional recurrence, while disease stage 3, positive axillary lymph nodes >3, and radiotherapy increase distant recurrence. Hormonal therapy reduced the distant recurrence. Pathological tumor size >2 cm, disease stage 3, positive axillary lymph nodes >3, and presence of LVI increased, while hormonal therapy and chemotherapy reduced death. Conclusions. Clinical characteristic reflecting tumor invasions increased locoregional recurrence, distant recurrence, or death, while hormonal therapy and chemotherapy reduced such risks. The effect of radiation remained inconclusive but may increase the risk of distant recurrence. [ABSTRACT FROM AUTHOR]

Objective. To reevaluate the diagnostic value of breast imaging in the diagnosis of breast cancer in areas where health resources are limited. Methods. Patients were women presenting with breast lumps in two university-affiliated tertiary hospitals, Thailand, during 2006 and 2010. Clinical data were abstracted from the breast cancer registration database and patient records.The diagnostic predictive ability of ultrasonography and mammography was obtained from logistic regression analysis and presented with areas under the receiver operating characteristics (AuROCs) curves. Results. Among 3129 breast lumps (3069 women), 854 were diagnosed with breast cancer by certified pathologists. Age and size of lumps alone already predicted cancer correctly in 77.45% (AuROC = 77.45). Additional ultrasonography increased the prediction to 96.22% (P < 0.001). Additional mammography also increased the prediction to 95.99% (P < 0.001). Performing both imaging modalities did not increase the prediction clinically (0.01%-0.24%). More accurate prediction (2.07%-2.21%) may be added by fine needle aspiration cytology (FNAC). Conclusions. Breast imaging is still valuable in settings where health resources are limited. Single breast imaging (only either ultrasonography or mammography) is adequate for cancer diagnosis. It is therefore unnecessary to perform both imaging modalities. Accuracy of the diagnosis may be improved by FNAC, if available. [ABSTRACT FROM AUTHOR]

Objective: To describe the pattern of disease progression and to describe locoregional recurrence, distant recurrence, and death rates in breast cancer patients after complete treatment., Material and Method: Medical records of women diagnosed with breast cancer at two university affiliated tertiary care hospitals in the Northern Thailand that had complete treatments between 2006 and 2010 were traced. Extracted key information included patient clinical profiles and documented recurrence of cancer The causes of death were verified from breast cancer case registration database, death certificates through The Ministry of Internal Affairs'civil registration, by direct telephone contact, or by distributed prepaid postcards., Results: Medical records of 829 women diagnosed with breast cancer without prior evidence ofdistant metastasis, and had complete recommended treatment were included. Six hundred thirty seven women had not experienced any events up to the end of the follow-up (76.8%). The first occurring events were focused and categorized into three distinct types, locoregional recurrence (n = 83, median follow-up time = 34.2 months), distant recurrence (n = 78, median follow-up time = 35.4 months), and death without any evidences of locoregional or distant recurrences (n = 12, median follow-up time = 36.7 months). Distant recurrence after locoregional recurrence was reported (n = 33). There were 109 patient who had died (breast cancer related death) up to the end of the follow-up (13.2%). The three types of consecutively occurring deaths were death after locoregional recurrence without any distant recurrences (n = 15), death after distant recurrence with locoregional recurrence (n = 21), and death after documenited distant recurrence without any locoregional recurrences (n = 61)., Conclusion: The trend was that the rate of the first occurring locoregional recurrence was slightly higher than that of distant recurrence, The death rate in patients without any recurrences was much lower than in those experiencing prior recurrences. The rates of disease progression from local recurrence to distant recurrence and to death were approximately 5 to 7 times faster in patients who had experienced earlierprogressions.

Two cases of fascioliasis gigantica from Northeast Thailand presenting with cholecystitis and gall stones were reported. Both cases complained of abdominal pain. On laparotomy the worms were found. In one case five worms were recovered during bile duct exploration and bile drainage; eggs were also revealed in the bile. In the other case of ectopic fascioliasis one young adult worm was found in a nodule which adhered to liver and diaphragm. The parasites were identified as Fasciola gigantica.