Combining hyperspectral imaging techniques with deep learning to aid in early pathological diagnosis of melanoma.

BACKGROUND: Cutaneous melanoma, an exceedingly aggressive form of skin cancer, holds the top rank in both malignancy and mortality among skin cancers. In early stages, distinguishing malignant melanomas from benign pigmented nevi pathologically becomes a significant challenge due to their indistinguishable traits. Traditional skin histological examination techniques, largely reliant on light microscopic imagery, offer constrained information and yield low-contrast results, underscoring the necessity for swift and effective early diagnostic methodologies. As a non-contact, non-ionizing, and label-free imaging tool, hyperspectral imaging offers potential in assisting pathologists with identification procedures sans contrast agents. METHODS: This investigation leverages hyperspectral cameras to ascertain the optical properties and to capture the spectral features of malignant melanoma and pigmented nevus tissues, intending to facilitate early pathological diagnostic applications. We further enhance the diagnostic process by integrating transfer learning with deep convolutional networks to classify melanomas and pigmented nevi in hyperspectral pathology images. The study encompasses pathological sections from 50 melanoma and 50 pigmented nevus patients. To accurately represent the spectral variances between different tissues, we employed reflectance calibration, highlighting that the most distinctive spectral differences emerged within the 500-675 nm band range. RESULTS: The classification accuracy of pigmented tumors and pigmented nevi was 89% for one-dimensional sample data and 98% for two-dimensional sample data. CONCLUSIONS: Our findings have the potential to expedite pathological diagnoses, enhance diagnostic precision, and offer novel research perspectives in differentiating melanoma and nevus.

Identifying Chinese adolescents with a high suicide attempt risk.

Adolescent suicide has become a serious public health problem in China. Since suicide attempts are considered the strongest predictor of suicide completion, it is valuable to identify adolescents with a high suicide attempt risk. This study aimed to develop a decision tree model for the interactive prediction of high suicide attempt risk in Chinese adolescents. A classification tree analysis was conducted in a sample of senior high school students (N = 6,686) based on the CRUISE program. The results indicated that depression, anxiety, social support, gender, self-esteem, family cohesion and adaptability were significant predictors of high suicide attempt risk, and interactions among these predictors constructed a hierarchical decision tree model. The tree model offered a series of reliable rules to identify Chinese adolescents with high suicide attempt risk, for example, adolescents with high depression scores had the highest probability (69.22%) of having a suicide attempt, female adolescents with low social support and low depression scores had the second highest probability (57.58%), and adolescents with low anxiety, low family adaptability, and medium depression scores had the third highest probability (55.77%). These exploratory findings suggested that different screening criteria are needed to detect at-risk Chinese adolescents with different severities of depression.

[Antithrombin III for early diagnosis of DIC in sepsis patients: a retrospective analysis with 445 patients].

OBJECTIVE: To investigated the role of antithrombin III (AT-III) levels in the early diagnosis of disseminated intravascular coagulation (DIC) in patients with sepsis and the predictive effect of AT-III on the development of DIC. METHODS: A retrospective study was conducted. Patients admitted to intensive care unit (ICU) of the First Affiliated Hospital of China Medical University from January to December in 2015 were enrolled. The patients were divided into sepsis group and non-sepsis group according to the diagnostic criteria of sepsis. In addition, sepsis patients were divided into 3 subgroups according to the international society on thrombosis and haemostasis (ISTH) scores on the first day: overt DIC (ISTH ≥ 5), non-overt DIC (ISTH 1-4) and none DIC group (ISTH = 0). Blood routine test, prothrombin time (PT), fibrinogen (Fib), D-dimer, fibrin degradation products (FDP), acute physiology and chronic health evaluation II (APACHE II) scores, sequential organ failure assessment (SOFA) scores and ISTH scores were recorded on the first ICU day. AT-III was recorded during 7 days. The differences were compared among these 3 groups. Correlations of AT-III with various parameters were calculated by using Pearson correlation analysis in sepsis group and overt DIC group. Receiver operating characteristic (ROC) curves for diagnosis of DIC with AT-III, AT-III+PT were drawn to evaluate the diagnostic efficiency. The AT-III levels of DIC patients were compared between early-onset DIC and late-onset DIC during their ICU stay. The change of AT-III levels with time and prognosis in patients with early-onset DIC was compared between groups. RESULTS: Totally 445 patients were recruited, with 138 patients in sepsis group, and 307 in non-sepsis group. There were 20 patents diagnosed with overt DIC on the first ICU day, 115 patients non-overt DIC and 3 patients of none DIC. Twenty-five sepsis patients were diagnosed overt DIC during the ICU days. AT-III level in sepsis patients on the first ICU day were lower than that in non-sepsis patients [(55.29±13.92)% vs. (76.54±12.31)%, P < 0.01]. Patients with overt DIC had a lower AT-III level than non-overt DIC or none DIC patients [(43.85±13.00)% vs. (56.95±13.03)%, (68.00±16.52)%, both P < 0.05]. It was shown by Pearson correlation analysis that AT-III level of sepsis patients on the first ICU day was negatively correlated to ISTH score and PT (r value were -0.467, -0.654, both P < 0.01). AT-III level of overt DIC patient on the first ICU day was negatively correlated with PT (r = -0.675, P = 0.001). It was shown by ROC curve that area under ROC curve (AUC) of AT-III combined with PT for diagnosis overt DIC in sepsis patients was higher than that of AT-III or PT alone (0.843 vs. 0.763, 0.834), the sensitivity 90.0%, specificity 73.7%. The cut-off value for overt DIC diagnosis in sepsis patients of AT-III level alone was 48.5%, sensitivity was 78.0%, specificity was 70.0%. On the first ICU day, AT-III level was risen when ISTH score improved in patients with sepsis. There was similar change of AT-III level between patients with early-onset DIC and late-onset DIC. AT-III level increased with DIC improvement. CONCLUSIONS: AT-III level can be used for diagnosing sepsis-associated overt DIC independently or with a combination of PT. When ISTH score improved, AT-III level was risen in patients with sepsis associated DIC.