Artificial neural networks improve the accuracy of cancer survival prediction.

BACKGROUND: The TNM staging system originated as a response to the need for an accurate, consistent, universal cancer outcome prediction system. Since the TNM staging system was introduced in the 1950s, new prognostic factors have been identified and new methods for integrating prognostic factors have been developed. This study compares the prediction accuracy of the TNM staging system with that of artificial neural network statistical models. METHODS: For 5-year survival of patients with breast or colorectal carcinoma, the authors compared the TNM staging system's predictive accuracy with that of artificial neural networks (ANN). The area under the receiver operating characteristic curve, as applied to an independent validation data set, was the measure of accuracy. RESULTS: For the American College of Surgeons' Patient Care Evaluation (PCE) data set, using only the TNM variables (tumor size, number of positive regional lymph nodes, and distant metastasis), the artificial neural network's predictions of the 5-year survival of patients with breast carcinoma were significantly more accurate than those of the TNM staging system (TNM, 0.720; ANN, 0.770; P < 0.001). For the National Cancer Institute's Surveillance, Epidemiology, and End Results breast carcinoma data set, using only the TNM variables, the artificial neural network's predictions of 10-year survival were significantly more accurate than those of the TNM staging system (TNM, 0.692; ANN, 0.730; P < 0.01). For the PCE colorectal data set, using only the TNM variables, the artificial neural network's predictions of the 5-year survival of patients with colorectal carcinoma were significantly more accurate than those of the TNM staging system (TNM, 0.737; ANN, 0.815; P < 0.001). Adding commonly collected demographic and anatomic variables to the TNM variables further increased the accuracy of the artificial neural network's predictions of breast carcinoma survival (0.784) and colorectal carcinoma survival (0.869). CONCLUSIONS: Artificial neural networks are significantly more accurate than the TNM staging system when both use the TNM prognostic factors alone. New prognostic factors can be added to artificial neural networks to increase prognostic accuracy further. These results are robust across different data sets and cancer sites.

Urgency of medical care. Contrasting perceptions of HIV-positive patients and physicians.

BACKGROUND: To assess the agreement between patients and physicians about the appropriate time to seek medical attention for symptom complexes commonly encountered in human immunodeficiency virus (HIV) disease, identifying potential suboptimal utilization of health care services. METHODS: A questionnaire consisting of 25 clinical problems commonly encountered in the ambulatory care of patients with HIV infection was developed to survey opinions regarding the most appropriate time to seek medical attention. Participants included 70 anonymous HIV-positive patients attending a health department's early intervention clinic and 104 physicians recruited at academic conferences in 1992. RESULTS: Clinically and statistically (P < .05 after Bonferroni correction) significant disagreement between physicians and HIV-positive patients regarding perceived urgency to seek medical care was found in 22 of the 49 possible responses. In seven of these scenarios, patients perceived greater urgency to seek care than physicians, especially for relatively discrete complaints, such as oral lesions, lymphadenopathy, and Kaposi's sarcoma. Conversely, in 15 scenarios, physicians were more concerned than patients, especially for serious complaints, such as those associated with meningitis, retinitis, sinusitis, pneumonia, infectious diarrhea, and urinary tract infection. CONCLUSIONS: Substantial differences regarding the urgency to seek care exist between physicians and HIV-positive patients. Patients focused on the physical aspects of their disease and had difficulty appreciating important symptom complexes associated with serious but potentially reversible conditions. A need exists for health professionals to educate patients about the appropriate urgency to seek care for a number of common symptom complexes.

Nutrition education curriculum at the University of Nevada School of Medicine.

This article describes a developing nutrition education curriculum for the University of Nevada School of Medicine, a small and progressive, community-based medical school. The curriculum has been planned to reflect the longitudinal, interdisciplinary, and comprehensive integration of nutrition into new required and existing courses so that timely delivery of knowledge and development of skills will occur in the basic science years followed by direct application in the clinical clerkships. The availability of the "Nutrition and Cancer Education Objectives" is timely and provides overall direction and curricular goals for the simultaneous development of nutrition-cancer education. The coordination of the four-year curriculum of the entire medical school can be specifically targeted for the incorporation of this needed information synergistically at a most opportune time.

Acetazolamide and dexamethasone in the prevention of acute mountain sickness.

We randomly assigned 32 healthy backpackers to receive placebo, acetazolamide (250 mg twice a day), dexamethasone acetate (4 mg four times a day), or both drugs in combination to determine the drug efficacy in preventing acute mountain sickness (AMS) at altitudes of 3,650 to 4,050m (12,000 to 13,300 ft). The incidence of AMS was high but symptoms were generally mild. Combined drug therapy was superior to both placebo and single drug therapy in risk reduction. Using acetazolamide alone was moderately beneficial in preventing the occurrence of AMS, although minor side effects were frequent. The use of dexamethasone alone did not significantly reduce the AMS incidence, and discontinuing its use resulted in symptoms suggestive of adrenal insufficiency. For recreational backpackers, routine drug prophylaxis is not recommended, in view of the mild nature of this illness and the adverse effects of medications. The efficacy of combined acetazolamide-dexamethasone therapy warrants further investigation at higher altitudes, where AMS is more severe, and the dexamethasone should be withdrawn gradually to avoid a possible adrenal crisis.

Medical recommendations for wilderness travel. 3. Medical supplies and drug regimens.

Even if the wilderness traveler is fit and up-to-date on management of the many illnesses and injuries that may occur on an outing, proper care cannot be given unless the appropriate supplies are available. Thus, the contents of a medical travel kit should be carefully chosen, using the type of trip, duration of travel, and distance from medical care as criteria.

Medical recommendations for wilderness travel. 2. Field management of illness and injury.

Knowledge of basic treatment of several illnesses and injuries common in wilderness travel can save time and discomfort on the trail. In some cases evacuation and medical attention will be necessary, but proper initial measures can be crucial to a successful outcome.

Stress Fracture Diagnosis by Computer-Assisted Thermography.

In brief: To compare infrared thermography, radionuclide bone scanning, and routine x-ray in diagnosis, a prospective blind study was conducted on 17 athletes (ten men, seven women) who had exertional leg pain compatible with stress fracture. The athletes with 'hot' patterns were unable to resume their original activity, while those with 'cold' patterns were able to resume activity. Additionally, specific soft-tissue syndromes were identified. The data indicate that thermography may be an accurate, noninvasive new way to distinguish stress fracture from other causes of shinsplint syndromes in runners. Applied repeatedly, this technique could be used to follow the course of bone healing, permitting the earliest safe return to activity.