Deep learning-level melanoma detection by interpretable machine learning and imaging biomarker cues.

SIGNIFICANCE: Melanoma is a deadly cancer that physicians struggle to diagnose early because they lack the knowledge to differentiate benign from malignant lesions. Deep machine learning approaches to image analysis offer promise but lack the transparency to be widely adopted as stand-alone diagnostics. AIM: We aimed to create a transparent machine learning technology (i.e., not deep learning) to discriminate melanomas from nevi in dermoscopy images and an interface for sensory cue integration. APPROACH: Imaging biomarker cues (IBCs) fed ensemble machine learning classifier (Eclass) training while raw images fed deep learning classifier training. We compared the areas under the diagnostic receiver operator curves. RESULTS: Our interpretable machine learning algorithm outperformed the leading deep-learning approach 75% of the time. The user interface displayed only the diagnostic imaging biomarkers as IBCs. CONCLUSIONS: From a translational perspective, Eclass is better than convolutional machine learning diagnosis in that physicians can embrace it faster than black box outputs. Imaging biomarkers cues may be used during sensory cue integration in clinical screening. Our method may be applied to other image-based diagnostic analyses, including pathology and radiology.

The rate of primary hypothyroidism in diabetic patients is greater than in the non-diabetic population. An observational study.

BACKGROUND AND OBJECTIVE: The aim of our study was to identify the rate of diabetic patients treated for hypothyroidism and compare them with a group without type 2 diabetes mellitus (T2DM). PATIENTS AND METHODS: We reviewed the computerized clinical records of 5161 patients. We identified diabetic patients treated with l-thyroxine. We compared the prevalence of PH with those patients under treatment with levothyroxine without T2DM. We excluded patients with a thyroid neoplasia, thyroid surgery, panhypopituitarism, or surgical complications of multinodular goiter or a thyroid nodule. Subclinical hypothyroidism was not considered. RESULTS: We included 1848 adult patients with T2DM in the study group, 58% women and 42% men. For the control group, we included 3313 non-diabetic patients, 55% women and 45% men. The mean age in the study group was 52±7 years, and 47±4 years in the control group (p<.001). The rate of hypothyroidism in the study group was 5.7%, and in the control group 1.8% (odds ratio of 3.45; 95% confidence interval 2.51-4.79) (p<.001). CONCLUSION: A strong association between T2DM and hypothyroidism was found. We recommend a thyroid profile in all patients with T2DM, similar to the recommendation in type 1 diabetes mellitus.

Rosiglitazone as an option for patients with acromegaly: a case series.

INTRODUCTION: In the patient with acromegaly, pituitary surgery is the therapeutic standard. Despite undergoing surgery, a significant number of patients with acromegaly continue to have uncontrolled growth hormone secretion. These patients require other treatments such as external irradiation and/or drug therapy. CASE PRESENTATION: We present the clinical and laboratory responses to six months of treatment with rosiglitazone in four cases. In all four cases, the patients had persistent growth hormone overproduction despite previous surgical treatment and other conventional therapy. Case 1 is a 57-year-old Caucasian woman, case 2 is a 51-year-old Hispanic man, case 3 is a 32-year-old Hispanic woman, and case 4 is a 36-year-old Hispanic man. In three of these patients, basal and nadir growth hormone and insulin-like growth factor 1 levels were significantly decreased (P < 0.05 and P < 0.01, respectively). CONCLUSION: Rosiglitazone could be a treatment option in select patients with acromegaly.