Classification of breast masses in mammograms using genetic programming and feature selection.

Mammography is a widely used screening tool and is the gold standard for the early detection of breast cancer. The classification of breast masses into the benign and malignant categories is an important problem in the area of computer-aided diagnosis of breast cancer. A small dataset of 57 breast mass images, each with 22 features computed, was used in this investigation; the same dataset has been previously used in other studies. The extracted features relate to edge-sharpness, shape, and texture. The novelty of this paper is the adaptation and application of the classification technique called genetic programming (GP), which possesses feature selection implicitly. To refine the pool of features available to the GP classifier, we used feature-selection methods, including the introduction of three statistical measures--Student's t test, Kolmogorov-Smirnov test, and Kullback-Leibler divergence. Both the training and test accuracies obtained were high: above 99.5% for training and typically above 98% for test experiments. A leave-one-out experiment showed 97.3% success in the classification of benign masses and 95.0% success in the classification of malignant tumors. A shape feature known as fractional concavity was found to be the most important among those tested, since it was automatically selected by the GP classifier in almost every experiment.

Genetic programming and feature selection for classification of breast masses in mammograms.

A dataset of 57 breast mass mammographic images, each with 22 features computed, was used in this investigation. The extracted features relate to edge-sharpness, shape, and texture. The novelty of this paper is the adaptation and application of genetic programming (GP). To refine the pool of features available to the GP classifier, we used five feature-selection methods, including three statistical measures -- Student's t-test, Kolmogorov-Smirnov Test, and Kullback-Leibler Divergence. Both the training and test accuracies obtained were above 99.5% for training and typically above 98% for testing.

The ratio of 2nd to 4th digit length: a predictor of sperm numbers and concentrations of testosterone, luteinizing hormone and oestrogen.

The differentiation of the urinogenital system and the appendicular skeleton in vertebrates is under the control of Hox genes. The common control of digit and gonad differentiation raises the possibility that patterns of digit formation may relate to spermatogenesis and hormonal concentrations. This work was concerned with the ratio between the length of the 2nd and 4th digit (2D:4D) in humans. We showed that (i) 2D:4D in right and left hands has a sexually dimorphic pattern; in males mean 2D:4D = 0.98, i.e. the 4th digit tended to be longer than the 2nd and in females mean 2D:4D = 1.00, i.e. the 2nd and 4th digits tended to be of equal length. The dimorphism is present from at least age 2 years and 2D:4D is probably established in utero; (ii) high 2D:4D ratio in right hands was associated with germ cell failure in men (P = 0.04); (iii) sperm number was negatively related to 2D:4D in the right hand (P = 0.004); (iv) in men testosterone concentrations were negatively related to right hand 2D:4D and in women and men LH (right hand), oestrogen (right and left hands) and prolactin (right hand) concentrations were positively correlated with 2D:4D ratio and (v) 2D:4D ratio in right hands remained positively related to luteinizing hormone and oestrogen after controlling for sex, age, height and weight.

The relationship between breast asymmetry, breast size and the occurrence of breast cancer.

Breast cancer is the second most common cancer among women in the world and in developed countries it is the most common. The early identification of women at risk is therefore of great importance and any additional measures which may aid diagnosis, particularly in high risk groups, would be of benefit. Breast volume and breast asymmetry were calculated from mammograms of 250 women with breast cancer and compared with those of 250 age-matched controls. There was evidence that breast cancer patients had more breast asymmetry and larger breasts than age-matched healthy women. The former observation is the first evidence that high breast asymmetry may be a risk factor for breast cancer. Breast asymmetry is likely to be a predictor of, rather than the effect of breast cancer.

Symmetry and ovulation in women.

Women have cryptic ovulation, and self-observation methods of determining the timing of ovulation, such as monitoring cervical mucus symptoms or recording basal body temperature, are not reliable. It has recently become apparent that the symmetry of paired soft tissue traits, such as breast size and digit length, changes across the menstrual cycle. This paper presents evidence that symmetry in four paired soft tissue traits showed a marked increase on the day of ovulation. The difference (i.e. the asymmetry) between the size of the left and right trait in ears, 3rd, 4th and 5th digits was measured. The timing of ovulation was confirmed by real-time pelvic ultrasonography and trait measurements were made without knowledge of scan results. Asymmetry was lowest on the day of ovulation (day 0), decreasing by about 30% from day -1, and significant within-subjects changes occurred from days -2 to day -1, and day -1 to day 0. The practical and evolutionary implications of these findings are discussed.