Yes.From studying 800 white-collar workers over age 60, we have good data showing ultrasound can be useful in early detection of prostate cancer.

When biopsy shows cancer is there, ultrasound is about 90 percent correct in calling it cancer, while digital exams are only about 50 percent correct. So for every two cancers detected by ultrasound, digital exams generally find only one.

More important, ultrasound spotted 100 percent of the "potentially curable" lesions 1.5 centimeter or less in diameter. Digital exams found only 45 percent.

A negative digital exam was only 65 percent accurate in ruling out a cancer, while ultrasound was 85 percent correct.

Ultrasound is cost-effective. With our first 400 patients, the detection cost per case found was about $7,000. With cervical cancer, by comparison, it's around $8,500. And it costs up to $25,000 to find one case of breast cancer.

Every year, 96,000 cases of prostate cancer are diagnosed and 26,000 die from it. It's now men's third-leading cause of death from cancer.

Our problem today is that we don't diagnose the disease early enough. For eight out of 10 men who come to a urologist, it's already too late. Recent studies indicate that ultrasound can detect most curable lesions much earlier.

Transrectal ultrasound has taken a quantum leap in the past two years. Detractors are basing their judgments on the technology of the '70s.

Right now, the urologists are saying ultrasound is good for staging, for directing the biopsy needle and for following patients who still have their prostate after treatment. If it can do all that, what's to prevent it from finding cancer before they can feel it? -- Dr. Fred Lee

associate radiologist, St. Joseph Mercy Hospital, Ann Arbor, Mich.

No.Prostate cancer clearly is a major disease that we want to diagnose and cure, and ultrasound may be just what we need to do it -- but it has to prove its value.

Yes, the technology has improved in recent years. But it hasn't been proved ready for use in routine screening.

Two things must be shown first: that ultrasound can find a lot of the tumors you don't feel in a physical exam, and that finding them changes what happens to people.

Nobody's done that yet. And it will be a difficult thing to do, because it might require you to not treat some of those people. If you find a small tumor, how are you going to know that that person would ever die from it? If you treat him and he doesn't die, that doesn't mean you helped him.

I don't think anyone would deny that you can diagnose some cancers with ultrasound that you can't feel. The question is, how many? And does it make a difference whether you treat them now, or maybe next year when you do feel them? Sensitivity isn't as important as predictive value.

At autopsy, at least one man in three has prostate cancer. If ultrasound were great, it would find them all. Dr. Lee's data certainly haven't shown that.

The notion that we ought to screen for prostate cancer because we routinely screen for breast cancer doesn't hold up. Prostate cancer is less common. It's diagnosed during life in one of every 17 to 21 men, compared with one of 11 women for breast cancer. And breast cancer is much more likely to be life-threatening.

How much is society willing to spend to diagnose something? You could do a CT scan on everybody. You'd find a lot of cancers -- but would it make a difference? -- Dr. Gerald Chodak

co-director of urologic oncology and director, Prostate Cancer Screening Program, University of Chicago