Summary
Background
Deep convolutional neural networks (DCNNs) can classify skin diseases at a level equivalent to a dermatologist, but their performance in specific areas requires further research.
...Objective
To evaluate the performance of a trained DCNN‐based algorithm in classifying benign and malignant lip diseases.
Methods
A training set of 1629 images (743 malignant, 886 benign) was used with Inception‐Resnet‐V2. Performance was evaluated using another set of 344 images and 281 images from other hospitals. Classifications by 44 participants (six board‐certified dermatologists, 12 dermatology residents, nine medical doctors not specialized in dermatology and 17 medical students) were used for comparison.
Results
The outcomes based on the area under curve, sensitivity and specificity were 0·827 95% confidence interval (CI) 0·782–0·873, 0·755 (95% CI 0·673–0·827) and 0·803 (95% CI 0·752–0·855), respectively, for the set of 344 images; and 0·774 (95% CI 0·699–0·849), 0·702 (95% CI 0·579–0·808) and 0·759 (95% CI 0·701–0·813), respectively, for the set of 281 images. The DCNN was equivalent to the dermatologists and superior to the nondermatologists in classifying malignancy. After referencing the DCNN result, the mean ± SD Youden index increased significantly for nondermatologists, from 0·201 ± 0·156 to 0·322 ± 0·141 (P < 0·001).
Conclusions
DCNNs can classify lip diseases at a level similar to dermatologists. This will help unskilled physicians discriminate between benign and malignant lip diseases.
What's already known about this topic?
Deep convolutional neural networks (DCNNs) can classify malignant and benign skin diseases at a level equivalent to dermatologists.
The lips are a unique feature in terms of histology and morphology.
Previous studies of DCNNs have not investigated tumours on specific locations.
What does this study add?
This study shows that DCNNs can distinguish rare malignant and benign lip disorders at the same rate as dermatologists.
DCNNs can help nondermatologists to distinguish malignant lip diseases.
What are the clinical implications of this work?
DCNNs can distinguish malignant and benign skin diseases even at specific locations such as the lips, as well as board‐certified dermatologists.
Malignant lip diseases are rare and difficult for less trained doctors to differentiate them from benign lesions.
This study shows that in dermatology, DCNN can help improve decision‐making processes for rare skin diseases in specific areas of the body.
Plain language summary available online
Aims/hypothesis The aim of this study was to confirm a link between mitochondrial dysfunction and type 2 diabetes. Materials and methods Cellular levels of mitochondrial proteins, cellular ...mitochondrial DNA content, and mitochondrial function and morphology were assessed by MitoTracker staining and electron microscopy, in white adipose tissue of 12-week-old male wild-type, obese (ob/ob), and diabetic (db/db) mice. Results Levels of mitochondrial proteins were found to be very similar in the livers and muscles of all the mice studied. However, levels were greatly decreased in the adipocytes of db/db mice, but not in those of the wild-type and ob/ob mice. Levels of mitochondrial DNA were also found to be considerably reduced in the adipocytes of db/db mice. MitoTracker staining and under electron microscopy revealed that the number of mitochondria was reduced in adipocytes of db/db mice. Respiration and fatty acid oxidation studies indicated mitochondrial dysfunction in adipocytes of db/db mice. Interestingly, there was an increase in mitochondria and mitochondrial protein production in adipocytes of db/db mice treated with rosiglitazone, an agent that enhances insulin sensitivity. Conclusions/interpretation Taken together, these data indicate that mitochondrial loss in adipose tissue is correlated with the development of type 2 diabetes.
To analyze the clinical features, outcomes including efficacy of treatment, and prognostic factors of patients with immunoglobulin D multiple myeloma (IgD MM).
Seventy-five patients diagnosed with ...IgD MM were selected from the Korean Myeloma Registry database (www.myeloma.or.kr).
Median age was 57 years and the main presenting features were bone pain (77%). Renal function impairment and hypercalcemia were present in 40 (53%) and 20 (27%) patients. Sixty-seven patients (89%) had lambda light chains. Forty-eight patients (64%) were of stage III by International Staging System. Twenty-six patients (53%) had chromosomal abnormalities mostly by conventional cytogenetics. Thirty-nine patients (54%) were treated with vincristine, adriamycin, and dexamethasone chemotherapy; the overall response rate (ORR) of 56%. Sixteen patients (22%) received first-line chemotherapy including new drugs (bortezomib or thalidomide), with an ORR of 81%. At a median follow-up time of 28.6 months, median overall survival (OS) was 18.5 months. Age, extramedullary plasmacytoma, del(13) or hypoploidy, serum β2 microglobulin level, and platelet count were significant prognostic factors for OS.
IgD MM is an aggressive disease that is usually detected at an advanced stage. Despite a positive initial response, survival after relapse was dismal. Intensive treatment strategies before and following stem cell transplantation may improve outcomes in younger patients.
Nail melanoma (NM) is an important differential diagnosis in patients with longitudinal melanonychia. However, diagnosis is often challenging as it is difficult to differentiate from other pigmented ...nail disorders. The main challenge for diagnosis is obtaining adequate nail matrix biopsy specimens for histopathological assessment. Furthermore, the histopathological changes in the early stages of NM are subtle and contribute to a delay in diagnosis and care. Therefore, the integration of clinical and histopathological analyses is essential. Clinical and dermoscopic features, such as a broadened width of asymmetric bands in an irregular pattern, with multicolour pigmentation, periungual pigmentation, and continuous growth, are features that support the diagnosis of NM. The essential histological features that must be assessed are cellular morphology, architectural features, melanocyte density, and inflammatory changes. The reported mutations in NMs were BRAF (0–43%), NRAS (0–31%), KIT (0–50%), NF1 (0–50%), and GNAQ (0–25%). Surgery is the primary treatment for NM. The recommended treatment for in situ or minimally invasive NM is functional surgery, but cases with suspected bone invasion should be treated with amputation. Targeted therapy and immunotherapy are indicated for advanced stages of NM. This review summarizes the updated guidelines for the diagnosis and treatment of NM.
The aims of this study were to investigate the outcomes of second salvage auto-SCT and to identify the impacts of a second auto-SCT compared with systemic chemotherapy alone on disease outcome. Data ...from 48 patients who underwent second auto-SCT were matched to 144 patients (1:3) who received systemic chemotherapy alone from the Korean Myeloma Registry. Groups were matched for nine potential prognostic factors and compared for treatment outcomes. The median age of matching-pairs at relapse was 55.5 years. A total of 156 patients (81%) received vincristine, doxorubicin and dexamethasone induction therapy before the first auto-SCT. Thirty-five patients (73%) in the second auto-SCT group received novel agent-based therapies before the second auto-SCT, and similar proportion in both groups received novel therapies after relapse of front-line auto-SCT. With a median follow-up of 55.3 months, patients who underwent a second auto-SCT had significantly better median OS (55.5 vs 25.4 months, P=0.035). In multivariate analysis for OS, <18 months time to progression after first auto-SCT, International Staging System III and salvage chemotherapy alone were independent predictors for worse OS. The outcomes of second auto-SCT appear to be superior to those of systemic chemotherapy alone. A randomized trial comparing both treatment strategies is required.
Influenza affects a considerable proportion of the global population each year, and meteorological conditions may have a significant impact on its transmission. In this study, we aimed to develop a ...prediction model for the number of influenza patients at the national level using satellite images and provide a basis for predicting influenza through satellite image data.
We developed an influenza incidence prediction model using satellite images and influenza patient data.
We collected satellite images and daily influenza patient data from July 2014 to June 2019 and developed a convolutional long short-term memory (LSTM)–LSTM neural network model. The model with the lowest average of mean absolute error (MAE) was selected.
The final model showed a high correlation between the predicted and actual number of influenza patients, with an average MAE of 5.9010 per million population. The model performed best with a 2-week time sequence.
We developed a national-level prediction model using satellite images to predict influenza incidence. The model offers the advantage of nationwide analysis. These results may reduce the burden of influenza by enabling timely public health interventions.